By Maria F. Gallo, PhD
Professor, Chair, and Associate Dean of Research, College of Public Health, Division of Epidemiology, The Ohio State University, Columbus
Among people discontinuing contraception to try to become pregnant, those who were obese had a longer time to pregnancy compared to those who were underweight or optimal weight.
Burger T, et al. Association of obesity with longer time to pregnancy. Obstet Gynecol 2022;139:554-560.
Obesity has been increasing steadily in the United States for decades. By 2020, about 31% of adult, reproductive-age women were obese.1 Obesity has been associated with difficulty in becoming pregnant and infertility (defined as the inability to become pregnant after at least one year of having sex without using contraception).2,3 This association holds even among people trying to achieve pregnancy with the use of in vitro fertilization.4
Burger and colleagues extended this research by focusing specifically on people who had recently discontinued using contraception because they wanted to become pregnant. The authors used data from the Fertility After Contraceptive Termination (FACT) study, which was a study designed to evaluate the link between discontinuing intrauterine device (IUD) use and time to pregnancy. Participants in the FACT study were enrolled in 2011 to 2018 from clinics in six cities across the United States. To be eligible to join the study, individuals needed to be 18-35 years of age, sexually active with a male partner, want to be pregnant, and have discontinued a method of contraception (IUD, implant, pills, patch, ring, shot, barrier methods, or withdrawal) in the past 120 days. People were ineligible if they were known to be infertile, had been surgically sterilized, or were currently pregnant. Participants had their height and weight measured at enrollment.
With these measures, the authors calculated body mass index (BMI) and used the standard cut points to classify participants as underweight or optimal weight (lower than 25 kg/m2), overweight (25 kg/m2 to 29.9 kg/m2), or obese (30 kg/m2 or higher). Participants were asked to complete a telephone survey every six months for two years. The authors also reviewed their medical records to validate the date of any pregnancy. Pregnancies were dated using the best available information. Participants were asked at enrollment to report on their usual interval from the start of one period to the next during the past 12 months. From this, the authors classified participants as having regular menses (cycles of 21-35 days) or irregular menses (any other cycle length or “too irregular to say”).
The main analysis used Cox proportional hazard models to evaluate whether time to pregnancy from the day of contraception discontinuation differed for participants who were overweight or obese compared to those who were underweight or optimal weight. The authors evaluated a range of factors from the literature as possible confounders: age, race, Hispanic ethnicity, socioeconomic status, gravidity, current smoking or alcohol use, sexual frequency, and contraceptive method discontinued. From this list, they defined as confounders those factors that changed the association between the BMI categories and time to pregnancy by at least 10%. Based on this, three factors were confounders: race (white, Black, other), low socioeconomic status (yes, no), and contraceptive method that was discontinued (IUD, implant, shot, other). The authors decided a priori to include menstrual regularity (yes, no) as a confounder.
Of the 498 participants enrolled in the FACT study, 432 had pregnancy status measured for the first 12 months of follow-up and were included in the analysis. Their mean age was 28 years and most were white (54%), while the remainder were Black (38%) or other (8.5%). Overall, 34% were obese, 25% were overweight, and 41% were underweight or optimal weight. By one year after method discontinuation, 68.8% (95% confidence interval [CI], 64.3% to 73.2%) had experienced a pregnancy. Participants with obesity had a median of 8.2 months to pregnancy, while those who were underweight or optimal weight had a median of 5.3 months. This difference was statistically significant in the unadjusted analysis with a hazard ratio (HR) of 0.63 (95% CI, 0.48-0.83). Furthermore, this association remained after adjusting for race, low socioeconomic status, contraceptive method discontinued, and menstrual regularity (adjusted HR [aHR], 0.62; 95% CI, 0.44-0.89). Being underweight was associated with a longer time to pregnancy, but this was not statistically significant (aHR, 0.83; 95% CI, 0.59-1.17). The one-year pregnancy rate was 59.1% (95% CI, 51.0% to 67.4%) among those who were obese, 69.5% (95% CI, 60.5% to 78.1%) among those who were overweight, and 76.4% (95% CI, 69.7% to 82.6%) among those who were underweight or optimal weight.
COMMENTARY
The work by Burger and colleagues builds on a growing body of evidence that links obesity to longer time to pregnancy and infertility. The authors studied a convenience sample of people discontinuing contraception because they wanted to become pregnant. By detecting the association between obesity and longer time to pregnancy and infertility in a different population than those used in past studies, the present findings add support to the conclusion that an association exists.
A key strength of the present study was its prospective nature: Participants were enrolled and asked over time whether they became pregnant. This design reduces the likelihood of recall bias or selection bias occurring. Another strength was that participants had their height and weight measured at enrollment rather than relying on their
self-reports. This is important because often people overestimate their height and underestimate their weight.5 Finally, the study was strengthened by controlling for the type of contraceptive method that was discontinued recently. Except for surgical sterilization, contraceptive methods do not permanently affect fertility. However, methods could differ in their delay to fertility. A recent analysis pooled data from almost 18,000 participants from three large cohort studies of women trying to become pregnant. Those discontinuing injectable contraception required five to eight menstrual cycles to return to fertility.6 In contrast, the return to fertility was lower among those discontinuing the patch (four cycles), pill or ring (three cycles), and copper or hormonal IUD or implants (two cycles). Thus, it was important that the study by Burger and colleagues controlled for the type of method discontinued recently.
A primary limitation of the study is that Burger and colleagues did not have participants routinely test for pregnancy, but instead relied on the participant identifying the pregnancy. People with obesity seem to be at higher risk of spontaneous abortion.4 If participants differed in their timing and frequency of pregnancy testing, and thereby differed in their likelihood of identifying pregnancies that ended in spontaneous abortion, this could have affected their findings. Ideally, participants would have followed the same cadence for pregnancy testing during study follow-up. The authors considered including a measure of sexual frequency as a confounder, but because it did not appreciably affect the analysis, they ultimately did not adjust for this. The analysis would have been stronger if they could have adjusted for sexual frequency occurring on menstrual cycle days on which the participant was likely to be fertile.
Another key limitation was that Burger and colleagues enrolled participants who had discontinued contraception in the past 120 days and who were not pregnant at enrollment. They did not report the mean time since contraception discontinuation at enrollment and whether this differed by obesity status. People with the highest fecundity might have been ineligible for the study because they already were pregnant at the time of recruitment. This means that the one-year pregnancy rates were underestimations. This also could have biased the observed difference by obesity status. A limitation that the authors acknowledged was the potential for inaccurately identifying those with irregular menstrual cycles. Participants were asked to report on their usual cycle in the past 12 months, which could have been influenced by the contraceptive method used at that time. Finally, the study enrolled participants from six study sites but did not control for this in their analysis. Controlling for study site is standard practice, since the sites might have differed in their methodology used or in important unmeasured population characteristics.
Several possible mechanisms could explain the observed association between obesity and increased time to pregnancy. It could be that obesity directly interferes with ovulation. For example, fat cells produce estrogen. Excess estrogen can make the body react as if it were pregnant and, thus, prevent ovulation, causing amenorrhea or irregular menses. On the other hand, obesity could be associated with infertility without causing it. Polycystic ovary syndrome (PCOS) is one of the most common causes of infertility in women. PCOS also can cause obesity. In this way, obesity could be associated with infertility without being the cause of infertility. Future research should seek to better understand the mechanisms, since these are important for developing appropriate counseling and interventions for individual patients.
These findings suggest that providers should counsel those with obesity who are trying to get pregnant that they could experience more difficulty in achieving pregnancy than those with lower BMIs. They might benefit from more timely adoption of practices, such as monitoring their fertile days or using ovulation predictor kits for timed sexual intercourse. Preconception also could be a time in which people are highly motivated to focus on weight control. Obesity is associated with a range of pregnancy-related complications, including gestational hypertension, preeclampsia, and gestational diabetes, and offspring of women with obesity have higher risks of adverse outcomes, including metabolic syndrome and childhood obesity.7 Even modest weight loss can help improve reproductive health. However, to successfully help preconception patients manage their weight in a healthy way, providers must ensure that their clinical encounters do not contribute to the stigmatizing of weight, a practice that is widespread in our healthcare system and larger society.
REFERENCES
- Behavioral Risk Factor Surveillance System Data Updated. March of Dimes Peristats. www.marchofdimes.org/peristats
- Ramlau-Hansen CH, et al. Subfecundity in overweight and obese couples. Hum Reprod 2007;22:1634-1637.
- Gesink Law DC, et al. Obesity and time to pregnancy. Hum Reprod 2007;22:414-420.
- Zaadstra BM, et al. Fat and female fecundity: Prospective study of effect of body fat distribution on conception rates. BMJ 1993;306:484-487.
- Flegal KM, et al. Comparisons of self-reported and measured height and weight, BMI, and obesity prevalence from national surveys: 1999-2016. Obesity (Silver Spring) 2019;27:1711-1719.
- Yland JJ, et al. Pregravid contraceptive use and fecundability: Prospective cohort study. BMJ 2020;371:m3966.
- American College of Obstetricians and Gynecologists’ Committee on Practice Bulletins–Obstetrics. Obesity in pregnancy: ACOG Practice Bulletin, Number 230. Obstet Gynecol 2021;137:e128-e144.
