Introduction: Previous studies suggest that women who are underweight or obese have reduced fertility; however, there is a lack of epidemiological studies on the association between under- and overweight and parity (ie, the number of times a woman has given birth) within large diverse populations. The study was designed to determine the impact of body mass index (BMI) at age 20 on the likelihood of nulligravidity (never having become pregnant) and nulliparity (never having given birth).
Methods: The cross-sectional study involved 33,159 North American Adventist women aged 40 and older who were nulliparous at age 20 years. Women with a BMI of 20-24.9 were considered normal weight and served as the comparison group.
The women were presumed to have a healthy lifestyle (the Adventist Church encourages members to avoid smoking and drinking, and to adhere to a vegetarian diet), though the authors note that approximately half of members follow this healthy lifestyle.
Results: Of the overall group, 4,954 women (15%) were nulligravid and 7,461 (23%) women were nulliparous at age 40. As shown in the Table, the odds ratio for nulligravidity and nulliparity was significantly increased in women who at age 20 were underweight, overweight, or obese, with the greatest effect seen in women more-than-obese with a BMI of ≥32.5 kg/m2, after adjusting for age and marital status (Table). In addition, adjustment for menstrual irregularities and other potentially confounding factors did not markedly alter the relationships. The negative association between obesity on nulliparity was lessened among women who had already given birth to one child (OR, 1.6) and was negligible in obese women with two children.
|BMI <18.5 kg/m2||25-29.9 kg/m2||30-32.4 kg/m2||BMI ≥32.5 kg/m2||P value|
*Odds ratio adjusted for age and marital status using a BMI of 20-24.9 kg/m2 as the comparison group. See commentary for odds ratio adjusted for more potentially confounding factors.
Adapted from Jacobsen BK, Knutsen SF, Oda K, Fraser GE. Body maxx index at age 20 and subsequent childbearing: the Adventist Health Study-2. J Womens Health (Larchmt). 2013;22(5):460-466.
Conclusion: Being underweight and, to a greater extent, being obese early in adulthood is associated with an increased risk for nulliparity and nulligravidity later in life, even after adjusting for age, smoking, and menstrual irregularities.
Jacobsen BK, Knutsen SF, Oda K, Fraser GE. Body mass index at age 20 and subsequent childbearing: the Adventist Health Study-2. J Womens Health (Larchmt). 2013;22(5):460-466.
Commentary by Tamara L Wexler MD, PhD
Tamara L. Wexler, MD, PhD, is an endocrinologist specializing in reproductive and neuroendocrinology, and an Attending in Medicine, Massachusetts General Hospital, Boston, MA.
Jacobsen and colleagues use retrospective population analysis of a specific population to examine the relationship between BMI at age 20 and later pregnancy.
Women with pregnancy at <21 years old or at an unknown age were excluded from the analysis, as were those with outlier values of estimated BMI (<16.0 or >60.0 kg/m2). It is important to note that the authors used different BMI categories than those established by WHO groupings—for example, since only 0.8% of the included subjects reported obesity class II (BMI ≥35.0 kg/m2) at age 20, the authors lowered the highest-BMI classification threshold to ≥32.5 kg/m2. Reference was set as BMI 20-24.9 kg/m2.
The authors found that being underweight, overweight, and obese by reported BMI at age 20 were risk factors for later nulligravidity (never having become pregnant) and nulliparity (never having given birth). Using a BMI of 20.0-24.9 kg/m2 as reference, a BMI <18.5 kg/m2 at age 20 carried an odds ratio of 1.21 of nulligravidity and 1.15 nulliparity; a BMI 30-32.4 carried an odds ratio of 1.87 of nulligravidity and 1.67 nulliparity; and a BMI >32.5 carried an odds ratio of 3.12 of nulligravidity and 2.40 nulliparity (odds ratios adjusted for smoking, alcohol, ethnicity, education, menarche, and menstrual irregularities.)
The association was significant for underweight, overweight, and obese, but as seen by the odds ratios, it was stronger for overweight than underweight. BMI was not further divided by degree of thinness, so we cannot compare the effect of moderate to mild thinness on fertility measures.
The relative odds ratios for nulligravidity vs nulliparity at the different BMI levels suggest that overweight/obesity has a greater impact on the ability to become pregnant than on the ability to carry a pregnancy to term. The negative association with obesity was particularly pronounced for first pregnancy and birth (also seen in a study by Gesink Law et al in this issue of EndoScan). Odds ratios for nulligravidity and nulliparity were closer for underweight BMI.
The degree to which anovulatory cycles contribute to the results—whether from anorexia or underweight anovulation, or from PCOS—is not known. However, the authors did adjust for reported menstrual irregularities, which did not alter the relationship between weight and parity.
Overall, this large study suggests that weight, even during young adulthood, is associated with later fertility. Of note, it is not known to what extent BMI per individual was sustained over time. Limitations also include the self-reported nature of independent and dependent variables and lifestyle factors. BMI at age 20 was established by recall of the participants for their weight at the time, along with current height, not by direct measurements of weight or height at age 20. History of pregnancy and live birth was self-reported; pregnancy rate may thus be affected by recall, and by potential lack of awareness of pregnancies that subsequently ended in early miscarriage. Subject characteristics did differ between groups. Subjects were all members of the Adventist church, and thus generalizability may be hampered; by self-report, two-thirds of the subjects had never used alcohol. Results do not take into account the health (including weight) of male partners.
This study looked not only at obesity but also at underweight, finding a U-shaped relationship between BMI and fertility (nulliparity and nulligravidity.) It is important to underscore the potential impact of underweight (see also Bates et al, 1982) as well as of obesity on later fertility The suggestion that obesity in young adulthood may have repercussions on fertility in later life is particularly alarming given the high rates of adolescent and adult obesity in many countries.