PROJECT ABSTRACT
Reducing rates of maternal morbidity and mortality is a top public health priority in the United
States, but little is known about maternal health care utilization and health outcomes in the first
few years' postpartum, particularly for women with pregnancy complications. The first two years'
postpartum also coincides with the recommended minimum time period between delivery of one
infant and start of the subsequent pregnancy–the interpregnancy interval– and the
interconception period for women who become pregnant again. In addition, twenty-eight million
women (23%) in the US live in nonmetropolitan (rural) areas, and rural-urban disparities in
maternal health may exacerbate differences in health care use and outcomes for women with
pregnancy complications. Three of the most common (10-20% prevalence) pregnancy
conditions are hypertensive disorders of pregnancy, gestational diabetes, and prenatal
depression; each is associated with long-term chronic health conditions. Understanding how
pregnancy complications are associated with health care use and disease risk during the
recommended interpregnancy interval, and if these relations are moderated by rural-urban
residency, are critical gaps in the knowledge necessary to improve maternal health in the United
States. We propose to use 2006-2020 data from the Maine All Payer Claims Database, a
repository containing medical, and pharmacy claim data for Maine residents, which captures
approximately 11,000 deliveries per year. We will study maternal health care utilization
(emergency department visits, inpatient hospitalizations) and chronic disease diagnoses in the
first 24 months' postpartum among women with each of three pregnancy conditions
(hypertensive disorders of pregnancy, gestational diabetes, and prenatal depression) in
comparison to uncomplicated pregnancies, and evaluate whether associations differ by
residence in rural vs. urban areas. For women with pregnancy complications, we will also
estimate the impact of living in rural areas on short interpregnancy intervals (<24 months) and
the mediating effects of contraception initiation. Our statistical analysis will include descriptive
statistics and time-to-event analyses, using inverse probability of censoring weights to account
for loss to follow-up and quantitative bias analysis to account for potential misclassification and
residual confounding. Graduate students in the public health program will gain hands-on
research experience during all stages of this project. Strengths of our project include
longitudinal follow-up of women postpartum, regardless of payer, and using population-based
data from Maine, which has the highest percentage of residents living in a rural area (60%) in
the United States, making it ideal for examining the effects of rural residency. In addition, we are
including prenatal depression as one of our three pregnancy conditions, which is increasingly
being tied to chronic medical conditions among reproductive age women. The results of our
project may advance progress towards lowering maternal morbidity and mortality rates in the
US, particularly for women living in rural areas.