The effects of exercise on birth weight: a meta-analysis

American Journal of Health Studies, Winter, 2002 by Eddie T.C. Lam, Jill M. Black, Kathleen D. Little, Judith Ausherman, Codruta Rafiroiu

Abstract: Previous research regarding the effects of exercise on pregnancy outcomes has been inconsistent. The purpose of this study was to adopt an objective research technique, meta-analysis, to summarize and analyze different studies with controversial results. Three coding variables (i.e., exercise duration, pre-pregnancy weight, and age) were selected in this study for comparisons. Though the effect size (ES) of the pregnant mothers who exercised under 30 minutes was greater than that of those who exercised between 30 and 60 minutes, the overall ES (-107.838, 95% CI = -482.93 to 267.26) indicated no significant birth weight differences between the two groups. On the whole, the results of this study support the notion that a significant positive relationship exists between birth weight and the pre-pregnancy weight. However, regression analysis indicated that a significant p <. 05) positive relationship existed between birth weight and the pre-pregnancy weight of the pregnant mothers (R =.75, p =.04). The conclusions of this study should be cautiously interpreted since they are limited to the selected studies under investigation. Directions and recommendations for future research are discussed.

**********

Physiological factors such as cardiac output, oxygen consumption, hormones, etc. will alter when a person exercises. Such physiological variables also change during pregnancy (Bell & O'Neill, 1994; Gorski, 1985; Spatling, Fallenstein, Huch, Huch, & Rooth, 1992). The combined effects of exercise during pregnancy are a major concern for most medical care providers, educators, the general public, as well as pregnant women and their families. The most common question being raised is whether oxygen and nutrients are shunted from the uterus to the working muscles during exercise. As a consequence, will women deliver immature or underweight babies because of exercise activity during pregnancy?

Previous research studies have indicated that low birth weight (i.e., not more than 2500 g) babies were inferior to normal weight babies in reference to future growth and development. There appears to be a higher frequency of learning disabilities and behavioral problems (McCormick, Gortmaker, & Sobol, 1990; Ornstein, Ohlsson, Edmonds, & Asztalos, 1991; Sommerfeldt, Ellersten, & Markestad, 1996; Sommerfeldt, Troland, Ellersten, & Markestad, 1996; Szatmari, Saigal, Rosenbaum, Campbell, & King, 1990), lower Intelligence Quotient (Ornstein, Ohlsson, Edmonds, & Asztalos, 1991; Pharoah, Stevenson, Cooke, & Stevenson, 1994; Sommerfeldt, Ellersten, & Markestad, 1995), and motor clumsiness such as impaired balance (Largo et al., 1989; Marlow, Roberts, & Cook, 1989) among low birth weight babies.

The direct causes of low birth weight babies are not fully understood. However, many factors have been examined to determine their effects on pregnancy outcomes. These factors include supplements (e.g., Scholl & Johnson, 2000), nutrition (e.g., Breslin, 1998), and exercise (see Pivarnik, 1998 and Sternfeld, 1997 for a thorough review). Among these factors, the most controversial one seems to be the effects of exercise on pregnancy outcomes, particularly on the changes in birth weight. In their study, Bell, Palma, and Lumley (1995) compared women who were doing vigorous exercise prior to pregnancy and continued exercising during pregnancy with those who did not do regular vigorous exercise (controls). They found women who did more than 4 sessions (30 minutes each session) of vigorous exercise weekly at 25 weeks gestation had babies whose mean birth weight was 315g lower than their counterparts. Other studies also supported the idea that strenuous physical activity in pregnancy could be associated with low birth weight and earlier deliveries (Bell & O'Neill, 1994; Clapp & Dickstein, 1984).

Researchers also indicated that a number of other factors seem to be related to poor pregnancy outcomes, as well. These factors may include, but are not limit to, the following: (a) maternal weight prior and during pregnancy (e.g., Cnattingius, Bergstrom, Lipworth, & Kramer, 1998; Edwards, Hellerstedt, Alton, Story, & Himes, 1996; Kumari, 2001; Ogunyemi, Hullett, Leeper, & Risk, 1998), (b) maternal nutrition, such as nutrient deficiencies or toxicities, use of some herbal supplements, eating disorders (e.g., Brown & Kahn, 1997; Story & Alton, 1995), (c) socioeconomic issues, like poverty, low levels of education, limited availability of food (e.g., Jonas, Roder, & Chan, 1992; Otterblad-Olausson, Cnattingius, ex: Goldenberg, 1997), (d) lifestyle choices, such as smoking, alcohol or other drug use (e.g., Kallen, 2001; Korea, Pastuszak, & Ito, 1998), (e) age, for example, teens and women over age 35 (e.g., Gortzak-Uzan, Hallak, Press, Katz, & Shoham-Vardi, 2001; Seoud et al., 2002), (f) previous pregnancies, such as short intervals between pregnancies, poor prior pregnancy outcomes, multiple births (e.g., Meyer, Buescher, & Surles, 1999; Zhu, Haines, Le, McGrath-Miller, & Boulton, 2001), and (g) maternal health concerns, like high blood pressure, diabetes, or other chronic diseases (e.g., Lopez, Smith, & Gutierrez, 2002; Sibai, 1996; Xiong, Demianczuk, Saundersm, Wang, & Fraser, 2002).