Women with mild/moderate anemia who are seeking contraception will be randomized to a levonorgestrel (LNG) intrauterine system (IUS) or an LNG/ethinyl estradiol (EE)/ferrous fumarate combined oral contraceptive (COC) regimen and followed prospectively for 12-18 months. Approximately 400 participants will be enrolled. The primary hypothesis that the mean change in hemoglobin in the group assigned to the LNG IUS will be superior to the COC/ferrous fumarate (control) group after 12-months of product use.

Anemia continues to disproportionately affect marginalized women in resource-poor countries. In Africa/SE Asia, over 270 million women of reproductive age are anemic. Iron deficiency anemia causes 18% of maternal deaths worldwide. Multiprong approaches are needed to reduce the prevalence of anemia and the negative impact on family health.

Every menstrual cycle, women lose 14 to 19 mg of iron; this is iron that anemic women need to lead healthier lives. Though the relationships between iron loss from menstruation, absorption of dietary intake of iron, storing iron, and the impacts on hematologic parameters are complex, higher levels of menstrual blood loss are associated with lower hemoglobin values.

The levonorgestrel intrauterine system is a highly effective contraceptive product that also generally reduces menstrual blood loss. In research spanning over four decades, the product consistently raises hemoglobin levels and increases iron stores in broad populations of women, but particularly for women with heavy menstrual bleeding. This product is not widely available in resource-poor countries, due to higher costs relative to other contraceptives.

As a potential tool to alleviate anemia, the levonorgestrel intrauterine system has never been adequately tested. Previous research has never focused on anemic women, nor used proper scientific approaches to determine if the product can significantly increase hemoglobin and iron levels via reducing menstrual blood loss.

The overall goal of the proposed research is to give anemic women in Kenya an opportunity to try the levonorgestrel intrauterine system and with improved scientific approaches, measure the impact on hemoglobin and iron stores. In this randomized trial, the comparison product will be oral contraceptives containing iron supplement pills. If the levonorgestrel intrauterine system is found to work as hypothesized, then the product can become another tool to alleviate anemia among reproductive-aged women, resulting in healthier living and healthier beginnings to pregnancy when desired.

Inclusion Criteria:

• Provide sequential oral and written consents to screen for eligibility and enroll
• Female, aged 18-50
• Desire to use contraception or possibly switch methods
• Possession of a cell phone and willingness to be contacted
• Confirmatory low hemoglobin (minimum of 87 g/L and maximum of 126 g/L) taken 1 to 4 weeks after the initial test
• Regularly menstruating (at least once every 35 days)
• At least 6 weeks postpartum
• Willingness to agree to study procedures
• Willingness to be randomized to treatment
• Willingness to use assigned hormonal contraceptive for 18 months

Exclusion Criteria:

• Severe anemia (hemoglobin < 86 g/L)
• Currently pregnant based on urine pregnancy test and failure to rule out possible pregnancy according to WHO guidelines
• Surgically sterilized or had a hysterectomy
• Participated previously in this study by being randomized to contraceptive
• Currently using a subdermal contraceptive implant and does not wish to have it removed
• Currently using an LNG IUS
• Currently receiving treatment for anemia
• Any of the following previously diagnosed disorders: hemosiderosis, hemochromatosis, hemolytic anemia
• Known allergic reactions to oral contraceptives or LNG IUS
• Currently receiving an investigational (unapproved) drug in another study
• Any condition (social or medical) which in the opinion of the investigator would make study participation unsafe
• Intending to become pregnant in the next 18 months
• Mucopurulent cervicitis
• Pelvic inflammatory disease
• Other medical contraindications to treatment, according to WHO Medical Eligibility Criteria (categories 3 or 4)

Data generated under the project will be shared consistent with the NIH Grant Policy, and the FHI 360 Protection of Human Subjects (PHSC) committee guidelines. Data will be de-identified to avoid linkages to individual research participants and will be free of variables that could lead to deductive disclosure of the identity of individual research participants. A de-identified primary dataset will be posted to an open-data repository within 30 calendar days after the first publication.

Deidentified quantitative research data will be made available in a timely manner once the main findings from the research have been accepted for publication.

Data generated under the project will be shared consistent with the NIH Grant Policy, and the FHI 360 Protection of Human Subjects (PHSC) committee guidelines. Data will be de-identified to avoid linkages to individual research participants and will be free of variables that could lead to deductive disclosure of the identity of individual research participants. A de-identified primary dataset will be posted to an open-data repository within 30 calendar days after the first publication.