Research Summary

Donna Day Baird, PhD, a reproductive epidemiologist with research focused on fertility, hormones, pregnancy loss, and more recently uterine leiomyoma, a common condition with symptoms that adversely affect the quality of life for nearly a third of reproductive-age women. Baird’s research on these outcomes utilized innovative methods or study designs to lay the groundwork for subsequent investigation of the environmental factors that influence them. Dr. Baird formerly led the Women’s Health Group within the Epidemiology Branch. As Scientist Emerita, Baird continues to carry out research in these three areas:

• Fertility
• Hormones and pregnancy loss
• Uterine leiomyoma

Fertility

Fertility is a challenging reproductive outcome to study because most couples begin their pregnancies privately, and medical record data can be used only for the select subset who seek medical care when they have trouble conceiving. Baird examined a continuous measure of fertility, the number of months of unprotected intercourse before conception (Time To Pregnancy). These data can be collected from pregnant or previously pregnant women who were not using birth control when conceiving. She designed her first study to evaluate the ability to collect detailed data on this outcome and test the hypothesis that cigarette smoking delays conception. The outcome data could be collected, and smokers did require a significantly longer Time To Pregnancy compared to non-smokers, providing evidence for the adverse effects of smoking on fertility. Her additional Time To Pregnancy findings include adverse effect on fertility for dental assistants exposed to nitrous oxide, for women exposed prenatally to DES, and for obese men. With her NIEHS biostatics colleague, Clare Weinberg, both the benefits and limitations of this method were described for other researchers interested in studying fertility effects of other exposures. The method has subsequently been used to study numerous occupational and environmental exposures around the world.

Hormones and Pregnancy Loss

As a coinvestigator on the Early Pregnancy Study (designed to measure very early pregnancy loss before study participants even knew they were pregnant). Participants collected daily morning urine samples during their pregnancy attempt and into early pregnancy for those who conceived. Baird’s research focused on investigating hormonal variation in menstrual cycles and very early pregnancy by measuring urinary metabolites of estrogen and progesterone in the daily urines. No reliable assay for urinary luteinizing hormone was available at the time, but the algorithm Baird developed in collaboration with Weinberg identified a good surrogate for ovulation, the Day of Luteal Transition (DLT), which is marked by the drop in estrogen and initial rise of progesterone. A later study by others found DLT to be highly concordant with day of ovulation as determined by ultrasound.

An analysis of long menstrual cycles observed in the Early Pregnancy study (using DLT to identify ovulation) revealed that a delayed estrogen rise, the pattern seen during breastfeeding, was rare. Instead, patterns were heterogeneous with the most common being an early rise with prolongation of the rise so that final estrogen concentrations tended to be higher than in cycles of normal length. This pattern would be consistent with delayed selection of a dominant follicle or failure and replacement of an initial dominant follicle.

Pregnancy cycles in the study were identified by a very sensitive immunoradiometric assay for the pregnancy hormone, human chorionic gonadotropin, (hCG). Having a day of ovulation allowed the study team to learn details of early pregnancy biology. For example, most implantations occur 7-10 days after ovulation, and pregnancies with later implantations often failed because the corpus luteum was not maintained (indicated by decline in urinary progesterone). Among pregnancies that survived beyond the first six weeks, the pattern of hCG rise during the first week post implantation was unrelated to number of fetuses, risk of spontaneous abortion or sex of the baby, but pregnancies with late implantations (after 10 luteal days) had slower rates of hCG increase. Baird took the unique opportunity provided by having data on time of implantation to describe various patterns of corpus luteum rescue or failure as reflected by the daily urinary progesterone concentrations; those with corpus luteum rescue that showed a delayed rise in progesterone had significantly shorter gestations. Baird also collaborates on more traditional studies of pregnancy loss.

Uterine Leiomyoma

With a recognition of the limited research on women’s gynecologic conditions, Baird expanded her research to uterine leiomyoma (fibroids), a hormonally-dependent benign tumor that develops during reproductive-age. The aim of the first study was to measure fibroid prevalence in Black and White women to test the long-held belief that Black women were at greater risk. The study design involved reproductive aged participants identified from a prepaid health plan (>80% of each ethnic group enrolled). Standardized ultrasounds were conducted on those not previously diagnosed to identify any fibroid ≥0.5 cm in diameter. With statistical collaborator, (David Dunson, NIEHS) estimates of cumulative incidence by the end of reproductive years indicated that nearly 70% of whites and >80% for blacks develop fibroids. Data showed a nearly 10-year earlier age of onset for Blacks. Both exercise and adequate vitamin D were identified as likely protective factors. A second study designed to measure fibroid growth based on study MRIs at 6-month intervals indicated that fibroids within a given woman can be highly variable in growth dynamics, including some shrinkage.

For a clearer view of the disease incidence and progression, Baird initiated her current ongoing longitudinal study (Study of Environment, Lifestyle and Fibroids, SELF) to focus on Blacks, the more vulnerable group. It uses periodic ultrasound examinations to detect fibroid incidence and measure growth of existing fibroids and extensive data collection to identify risk factors (participants: n=1693 23-35 year olds enrolled 2010-2012 who live in the Detroit, Michigan area). Participants have undergone five standardized ultrasounds at the collaborating medical center and shared extensive life course data. Examples of the numerous published findings include one supporting the protective effects of vitamin D, several that found no support for the long-standing, but essentially untested hypothesis that reproductive tract infections are risk factors for fibroids, and two examining hormonal contraceptive exposure. In collaboration with extramural collaborators, endocrine disrupting chemicals measured in participant specimens are also being examined (individually and as mixtures) to identify associations that could point to modifiable exposures that could reduce risk.

Baird is an Adjunct Professor in the Epidemiology Department at the University of North Carolina at Chapel Hill and has mentored numerous masters and doctoral students in the program. She’s been a long-standing member of the American Society for Reproductive Medicine where she has participated in a Symposium on Uterine Fibroids and lectured for the Uterine Fibroid Interest Group. She was chosen for the American Public Health Association’s John Snow Award in 2023 “for distinguished service to the health of the public through outstanding contributions to epidemiology.”

Studies

Minnesota Time to Pregnancy Study

California Dental Assistants Study

Early Pregnancy Study

Uterine Fibroid Study

Study of Environment, Lifestyle & Fibroids