The Centers for Disease Control and Prevention (CDC) and the American College of Sports Medicine (ACSM) recommend 30 min or more of moderate physical activity on most, if not all, days of the week to improve the health and well being of all individuals (1,12). The CDC and ACSM also state that more intense physical activity performed in 20- to 60-min sessions on 3–5 d·wk−1 will result in higher levels of physical fitness. The American College of Obstetrics and Gynecology (ACOG) has presented additional recommendations that pregnant women should exercise with similar safeguards as nonpregnant women provided that there are no medical or obstetric complications during the pregnancy (1,3). Generally, women beginning an exercise program during pregnancy should perform moderate, non–weight bearing activities. Previously active women without medical or obstetric complications should continue their activities, except for contact sports, scuba diving, or other activities that might possibly cause abdominal distress (4).
Only a few studies have identified the frequency, duration, and type of recreational physical activities performed during pregnancy. In a cross-sectional study from the 1988 National Maternal and Infant Health Survey, 42% of women reported exercising during pregnancy and half of these women exercised for more than six months (16). Walking (43%), swimming (12%), and aerobics (12%) were the most frequent activities. In a Washington case-control study of 386 women from 1998 to 2000, 61% of pregnant women participated in some regular physical activity (11).
Previous analyses, however, have not focused upon the extent that these women are meeting the established recommendation levels for physical activity, and whether physical activity levels differ between pregnant and nonpregnant women. The goal of our study was to compare the physical activity levels and trends among pregnant and nonpregnant women based upon current recommendations. The importance of comparing pregnant and nonpregnant women is vital to understand who is meeting the recommendation levels and whether a difference exists in the frequency and duration of physical activity based upon pregnancy status.
We conducted a population-based, cross-sectional study using publicly available data from the 1994, 1996, 1998, and 2000 Behavioral Risk Factor Surveillance System (BRFSS). The data were utilized to determine the prevalence of physical activity among pregnant and nonpregnant women in the United States, in accordance to physical activity requirements previously established by the CDC and ACSM (12). The BRFSS data from 1994 to 2000 were obtained from monthly, year-round random-digit dialing telephone interviews in all 50 states, the District of Columbia, Puerto Rico, Guam, and the U.S. Virgin Islands. The data include the responses to core questions regarding physical activity, and represent the most recent population-based information regarding physical activity among pregnant and nonpregnant women in the United States. The CDC established the BRFSS in 1984 to collect information concerning behavioral risk factors, and it is currently the largest ongoing health surveillance system in the world (8,13).
Study population and algorithms.
Our study population included women who were 18–44 yr of age and were identified as having answered “yes” (pregnant) or “no” (not pregnant) to a question regarding the current knowledge of their pregnancy status. The pregnant and nonpregnant women were then categorized into mutually exclusive groups based upon their level and duration of physical activity in the past month. The groups consisted of vigorous activity (meeting guidelines: 20 min or more, three or more times per week at an intensity of ≥6 METs (metabolic equivalents)), moderate activity (meeting guidelines: 30 min or more, five or more times per week at an intensity of 3.0–5.9 METs), moderate or vigorous activity (not meeting guidelines: ≥150 min|b1wk−1, regardless of frequency at an intensity of ≥3.0 METs), irregular physical activity (not meeting any of the prior three recommendations), or no physical activity. The two most common activities (from a list of 56) performed in the month before the telephone interview were used to define the level and duration of physical activity. The appropriate MET values were assigned to the first and second activities based upon prior classification (2). The women in the study population were assigned to the highest category for which the recommendation was met in either or both activities.
The following sociodemographic covariates from the BRFSS were evaluated, as they have been associated with physical activity during pregnancy in prior studies: age, race, education, employment, income, marital status, and smoking status (11,16). Maternal age was coded as a categorical variable, including 18–24, 25–29, 30–34, and 35–44 yr. Race was categorized as non-Hispanic white, non-Hispanic black, Asian/pacific islander, American native, Hispanic, and other. Education was categorized as <12, 12, and >12 yr of education. Employment was coded as a categorical variable, including employed, not employed/student, and homemaker. Income was categorized as <$20,000, $20,000–$34,999, $35,000–$49,999, $50,000–$74,999, and ≥$75,000. Marital status was dichotomized as married or not married, and smoking status was categorized as current smoker, former smoker, and never smoked.
The prevalence of physical activity among pregnant and nonpregnant women was calculated using cross tabulations. Chi-square analysis was used to determine if there were significant differences in the sociodemographic covariates and physical activities for pregnant and nonpregnant women. A chi-square test of trend was used to determine if the changes in physical activity recommendation levels were significant over time.
Crude prevalence odds ratios (POR) and 95% confidence intervals (CI) were computed to determine the likelihood of physical activity among pregnant women. Crude POR were adjusted for all covariates in this study using logistic regression analysis to control for any confounding bias. SAS and SUDAAN were used to compute standard errors to account for the complex sampling scheme of the BRFSS data. Study participants categorized as having “no physical activity” served as the reference population for the logistic regression analysis.
The sample population included 150,259 women (6,528 pregnant and 143,731 nonpregnant) who had been surveyed for the BRFSS during the years of 1994, 1996, 1998, or 2000. This study was restricted to women 18–44 yr of age.
Table 1 shows the sociodemographic covariates of our study population. Differences in age, race, education, employment, income, marital status, and smoking status existed between pregnant and nonpregnant women. Pregnant women were generally younger than nonpregnant women. Sixty percent of pregnant women were 18–29 yr old, whereas 58% of nonpregnant women were 30–44 yr old. Most pregnant and nonpregnant women were non-Hispanic white (67 and 68%, respectively), had more than 12 yr of education (57 and 60%, respectively), and had an annual income less than $50,000 (60 and 60%, respectively). More pregnant women were homemakers as compared with nonpregnant women (25 and 15%, respectively). Seventy-two percent of pregnant women were married compared with 55% of nonpregnant women. Many pregnant and nonpregnant women had never smoked (66 and 61%, respectively); however, 12% of pregnant and 25% of nonpregnant women were current smokers.
Figure 1 shows the most common forms of physical activity among the women during 1994–2000. Walking was the most common physical activity among both pregnant and nonpregnant women (52 and 45%, respectively). Besides those women who were walking, there were notable differences between the percentages of pregnant and nonpregnant women performing aerobics (8 and 14%, respectively; P < 0.001) and running/jogging (2 and 7%, respectively; P < 0.001). Similar percentages of gardening (3 and 5%, respectively) and swimming laps (4 and 3%, respectively) were reported for both pregnant and nonpregnant women.
Table 2 displays the trends in physical activity among the study population from 1994 to 2000. Overall, nonpregnant women were more likely to meet the vigorous and moderate recommendations. The percentage of women meeting the vigorous recommendation has declined from 1994 (9% pregnant and 17% nonpregnant) to 2000 (6% pregnant and 14% nonpregnant); however, this trend was not significant (P = 0.10). There was no significant trend over time among the women meeting the moderate guideline for pregnant and nonpregnant women.
Adjusted POR are presented in Table 3. Pregnant women meeting the moderate or vigorous physical activity recommendations were more likely to be younger, non-Hispanic white, more educated, married, nonsmokers, and to have higher incomes. The likelihood of pregnant women meeting these recommendations increased with decreasing age and with increasing years of education and levels of annual income. Similar trends were seen for those meeting current recommendations for physical activity regardless of the intensity of the activity.
The physical activity recommendations met by pregnant women from 1994 to 2000 showed that a greater prevalence of these women were performing irregular activity or were not physically active, as compared to nonpregnant women in this study. These observations may imply that women are decreasing the duration, frequency, and intensity of physical activity once they become pregnant; however, this was not a longitudinal sample, so we were unable to determine physical activity changes as the women became pregnant.
More pregnant and nonpregnant women were walking, as compared with performing vigorous activities such as aerobics or running. Even though more women were participating in moderate activity, they did not achieve the frequency and duration requirements of 30 min or more, five or more times per week, or were active for 150 min or more per week. In comparison, among those women performing vigorous activities such as aerobics, approximately two thirds met the vigorous recommendation level.
Prior studies have evaluated the prevalence of physical activity during pregnancy and have shown varying frequency and duration measurements. Evenson et al. stated that the prevalence of recommended activity (moderate-intensity activity at least 5 times per week for at least 30 min each time or vigorous-intensity activity at least 3 times per week for at least 20 min each time or both) was lower among pregnant women than nonpregnant women (15.8 vs 26.1%) (6). Ning et al. noted that among 61% of women reporting some regular activity during pregnancy, more than half exercised less than 4 h|b1wk−1 (11). Approximately 27% of these pregnant women reported performing regular, high-intensity activities (≥6 METs) during the first 20 wk of pregnancy, and half were active more than 4 h|b1wk−1 (11). Zhang and Savitz reported that 35% of U.S. women exercised at least 3 times per week before and during pregnancy, 7% began an exercise program during pregnancy, 13% stopped exercising when they became pregnant, and 45% did not partake in regular exercise before or during pregnancy. About half of those who exercised during pregnancy were active greater than 6 months (16). Hinton and Olson indicated that 64% of white women exercised before pregnancy and then approximately 20% became more active and 40% became less active during pregnancy (9). Additionally, Mottola and Campbell specified that 70 and 49% of their study population participated in a structured exercise program before pregnancy and in the third trimester, respectively (10).
The prevalence of physical activity among pregnant women in our study is much lower in comparison with other studies, due to the frequency and duration requirements of the recommendation levels. Comparisons between studies of physical activity prevalence, however, should be made with caution, as different measurements were taken to assess activity.
Zhang and Savitz reported that women who exercised into their pregnancy decreased their average intensity and weekly duration of activity compared with 1 yr before their pregnancy (7.2 to 3.0 h|b1wk−1) (16). Our study as well implies that pregnant women performed less intense physical activities with less duration and frequency than nonpregnant women.
Walking has been the most widespread physical activity among pregnant women in earlier studies, is the most common form of moderate activity, and is accessible across socioeconomic strata (7,14). In our study, the prevalence of walking among pregnant women was similar (52%) to that reported elsewhere (60% by Ning et al. and 43% by Zhang and Savitz) (11,16). Reports of other common forms of physical activity during pregnancy included swimming, gardening, jogging, and aerobics. The occurrences of these activities were relatively similar among other studies, but higher in relationship to this study as more women were walking (11,16).
The sociodemographic covariates previously reported that are correlated with vigorous activities included younger age, increased education, and higher income (11). This study described similar associations between these covariates and the moderate and vigorous recommendation levels. This study was consistent with previous reports that smokers, married, and women of color were significantly less active than nonsmokers, unmarried, and non-Hispanic, white women, respectively (11). Our analysis, however, did not provide significant results to indicate that working women were more active than not working, as reported elsewhere (11,16).
This study largely illustrates that there appears to be a discrepancy between the professional recommendations for physical activity during pregnancy and what women have actually been doing. Even though physical activity recommendations for pregnant women were initially established in the 1980s, there was not a consistent body of evidence to support these guidelines (15). Moreover, this lack of evidence could provide insight as to why little information has been available to physicians and midwives to promote regular, moderate exercise during pregnancy. These healthcare providers may not have been convinced that exercise was safe during an uncomplicated pregnancy and consequently were not advising this activity (1).
However, in 2003, evidence-based statements reassuring healthcare providers of the safety of exercise during pregnancy were proposed with the Clinical Practice Guideline, published jointly by the Society of Obstetricians and Gynaecologists of Canada (SOGC) and the Canadian Society for Exercise Physiology (CSEP) (5). This research focused on the outcomes of aerobic and strength-conditioning exercises and did not find an increase in early pregnancy loss, late pregnancy complications, abnormal fetal growth, or adverse neonatal outcomes (5). In 2004, the ACSM endorsed and promoted the use of this guideline (http://www.acsm.org/sports_medicine_bulletin_master/SMB_395/SMB_395_admin_council.html).
The Clinical Practice Guideline also recommends the use of the “Physical Activity Readiness Medical Examination for Pregnancy,” published jointly by CSEP and Health Canada. This document describes the medical clearance for prenatal exercise participation and specific research-based advice for conditioning and safety (http://www.csep.ca/forms.asp). These safety procedures will further educate healthcare providers about the appropriate promotion of exercise during pregnancy.
Furthermore, obstetricians and gynecologists should focus on encouraging continued physical activity during pregnancy among those already active and should specifically target physical activity promotion among those women performing irregular or no activity. This target group would most likely consist of women who are either nonwhite, current smokers, or have low annual incomes or less than 12 yr of education. These women should be encouraged to begin moderate activities most, if not all days of the week, as long as medical or obstetric complications do not exist. Overall, this study has vital public health implications that can assist physicians to identify patients who are at high risk for inactivity during pregnancy.
There are specific limitations for this study. Data from the BRFSS were assessed by the use of self-reported measures, which are prone to differential bias. It is highly unlikely that all women were able to precisely evaluate their physical activity frequency and duration. Further, women were probably more apt to overreport their activity levels due to the appeal of exercise. Some activities, such as swimming, were categorized as being vigorous (6 METs); however, it is again improbable that all pregnant women swimming were exerting 6 or more METs. The BRFSS did not evaluate the woman’s time of gestation during which the questions were asked. This could have led to major differences in activity levels among the study subjects, as women are more likely to decrease the frequency and intensity of physical activity in the later stages of pregnancy (9). Our sample could not ascertain how many women had high-risk pregnancies and were advised by their doctors not to partake in physical activity.
Another limitation to this study involves the timeline for physical activity recommendations. Initial exercise recommendations in the 1980s were conservative (maximum pulse rate of 140 bpm and 15 min of aerobic activity) due to a lack of scientific evidence supporting aerobic exercise during pregnancy (15). The guidelines have become less restrictive over the past 20 yr due to evidence that has promoted safe prenatal exercise (15). ACOG’s support for the CDC’s recommendation of regular physical activity, however, was not published until 2002, which follows the data collection for our study (15). The older restrictive guidelines may have contributed to reductions in habitual activity during pregnancy.
Further studies should prospectively explore how commonly pregnant women are meeting the recommended levels of physical activity; collect detailed information about the pregnancy and the frequency, duration, and intensity of activities; and assess physicians regarding how well they promote and educate women about physical activity during an uncomplicated pregnancy.
1. ACOG Committee Opinion Number 267. Exercise during pregnancy and the postpartumperiod. Obstet. Gynecol.
2. Ainsworth, B. E., W. L. Haskell, M. C. Whitt, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc.
3. Artal. R., and M. O’Toole. Guidelines of the American College of Obstetricians and Gynecologists for exercise during pregnancy and the postpartum period. Br J Sports Med.
4. Artal. R., and C. Sherman. Exercise during pregnancy. Phys Sportsmed.
5. Davies, G. A. L., L. A. Wolfe, M. F. Mottola, and C. MacKinnon. Joint SOGC/CSEP clincal practice guideline. Exercise in pregnancy and the postpartum period. Can J. Appl. Physiol.
6. Evenson, K. R., D. A. Savitz, and S. L. Huston. Leisure-time physical activityamong pregnant women in the US. Paediatr. Perinat. Epidemiol.
7. Eyler, A. A., R. C. Brownson, S. J. Bacak, and R. Housemann. The epidemiology of walking
for physical activity in the United States. Med Sci Sports Exerc.
8. Gentry, E. M., W. D. Kalsbeek, G. C. Hogelin, et al. The Behavioral Risk Factor Surveys: design, methods, and estimates from combined state data. Am J Prev Med.
9. Hinton, P. S., and C. M. Olson. Predictors of pregnancy-associated change in physical activity in a rural white population. Matern. Child Health J.
10. Mottola, M. F., and M. K. Campbell. Activity patterns during pregnancy. Can. J. Appl. Physiol.
11. Ning, Y., M. A. Williams, J. C. Dempsey, T. K. Sorensen, I. O. Luthy. D. A. Luthy. Correlates of recreational physical activity in early pregnancy. J. Matern. Fetal Neonatal Med.
12. Pate, R. R., M. Pratt, S. N. Blair, et al. Physical activity and public health. A recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA
13. Remington, P. L., M. Y. Smith, D. F. Williamson, R. F. Anda, E. M. Gentry, and G. C. Hogelin. Design, characteristics, and usefulness of state-based behavioral risk factor surveillance: 1981-1987. Public Health Rep.
14. Siegel, P. Z., R. M. Brackbill, and G. W. Health. The epidemiology of walking
for exercise: implications for promoting activity among sedentary groups. Am. J. Public Health
15. Wolfe, M. F., and G. A. L. Davies. Canadian guidelines for exercise in pregnancy. Clin. Obstet. Gynecol.
16. Zhang, J. D., and A. Savitz. Exercise during pregnancy among US women. Ann. Epidemiol.