PROVIDENCE, R.I. [Brown University] — A new study finds that providing people who have recently given birth access to long-acting reversible methods of contraception, such as intrauterine devices and contraceptive implants, could help prevent them from unintentionally falling pregnant in the following months.

The study — which analyzed the effects of a 2012 Medicaid policy implemented in South Carolina — found that expanded access to particular forms of birth control were especially helpful in preventing unintended pregnancies among adolescents who had just given birth, giving them more control over their own futures.

“The ability to control whether and when you become pregnant is a basic human right, since pregnancy and childbirth have enormous implications for social and economic life trajectories,” said Maria Steenland, an assistant professor of population studies (research) at Brown University who is affiliated with the Population Studies and Training Center.

Steenland conducted the study, published on Friday, Feb. 5, in Health Affairs, along with three colleagues at Harvard University’s medical and public health schools.

Steenland said that in 2012, South Carolina’s Medicaid program became the first in the United States to reimburse hospitals for the provision of immediate postpartum long-acting reversible contraception (LARC). LARCs, which prevent pregnancy for extended periods of time without any effort on the patient’s part, include intrauterine devices, arm implants and hormonal injections.

Before the state enacted the policy, she said, patients who had just given birth and wanted immediate postpartum contraception had few options; LARCs were only available if they were willing to come back to the hospital for a separate outpatient procedure, and birth control pills are not considered medically safe to use early in the postpartum period. As a result, 59% chose to use no highly or moderately effective method of contraception; 22% chose a short-acting method, such as spermicide; and 13% chose sterilization, which is irreversible.

To examine the impact of the new policy, the study analyzed data on more than 150,000 Medicaid-insured South Carolina women between ages 12 and 50 who gave birth from 2010 to 2014. They found that in the medical facilities that began to offer LARCs after the policy change, use of highly effective contraception among postpartum patients shot up among adolescents, who are particularly vulnerable to closely spaced, higher-risk repeat pregnancies. Among Medicaid patients age 12 to 19, the rate of immediate postpartum LARC use increased by more than 6 percentage points between March 2012 and November 2014. Across all age groups, the total percentage of Medicaid patients who opted for postpartum LARCs nearly quadrupled in two and a half years. In some facilities, medical workers were providing LARCs to up to 20% of all postpartum patients.

“Contraceptive choice is based on many factors, such as side effects, reversibility and effectiveness,” Steenland said. “Our study shows that making these new contraceptive methods available can make it easier for patients to find a method that meets their needs and preferences, and ultimately it can give them more agency in deciding whether and when to become pregnant again.”

Less encouraging, Steenland said, was the researchers’ discovery that as of 2014, fewer than half of South Carolina facilities had begun to offer immediate postpartum LARCs to patients, despite the policy change.

“We found that few hospitals began offering immediate postpartum LARCs after the policy change, indicating that Medicaid reimbursement is only a first step to making these options available,” she said.

Steenland and her coauthors recommended that facilities and the state Medicaid program take further policy steps to make long-acting contraception more widely available — especially given that many studies, including one of their own, show that when LARCs are available and free, more patients opt to use them and rates of unintended pregnancy decrease.

The work was funded in part by a grant from the population dynamics branch of the National Institute of Child Health and Human Development (R03HD099428). Steenland was supported by the National Institutes of Health (training grant #T32 HD007338) and by other NIH support (grant #P2C HD041020). The Eric M. Mindich Research Fund for the Foundations of Human Behavior provided funding to acquire the data used in this study.