Abortion in the United States: Occurrence, trends, cross-state care-seeking and unmet need

Thesis (Master's)--University of Washington, 2016-08

Introduction: Nationally, the abortion rate has fallen over the last two decades. During the same period, the number restrictions and regulations on abortion have increased, especially after 2010. From 2001 to 2008 the percent of unintended pregnancies ending in abortion has fallen by 7 percentage points indicating a potential decrease in access to abortion services or decreased demand. Therefore, it is crucial to determine the drivers of these declines to tease apart the effect of access, restrictions, family planning, demographic and socioeconomic factors. Methods: This study utilizes data from the Alan Guttmacher Institute (AGI) alongside the Center for Disease Control’s (CDC) abortion surveillance data to estimate the trends and levels in state-level abortion by occurrence, age-standardized occurrence, and residence from 1990 to 2012. These values are then used to model resident in-state care-seeking for abortion services controlling for sociodemographic characteristics, access and restrictions. Counterfactual scenarios are constructed to measure unmet need. Results: In almost every state, the abortion ratio by occurrence decreased between 1990 and 2012 with the largest decreases seen in those states with the lowest ratios. Few differences were observed between the birth-rate standardized ratio to the crude estimated occurrence ratio. Over half of the residents in Wyoming, Mississippi and Missouri sought their abortions from other states in 2012. The estimated unmet need in 2012 was 70.5 abortions per 1,000 live births to women aged 10-49. The largest single impact for reducing this unmet need was predicted for eliminating restrictions in Medicaid funding for abortions. Conclusion: Despite an overall decrease in the abortion ratio over the past decade, the expected abortion ratio would be considerably higher controlling for socioeconomic, demographic and cultural factors if there were more providers, if providers were distributed so every woman had access to a provider in her county and given changes in legislation and abortion restrictions. The largest impact on reducing unmet need was observed for lifting Medicaid funding restrictions, suggesting a disproportionate level of unmet need among low income women. There is also a need for more standardized data on abortion reporting at the state level to monitor the effect of restrictions on access to abortion services and to bolster family planning services in areas with the most unmet need.