Villages in the comparison group will continue with the standard of care for contraceptive services. This includes health facilities offering contraceptive counseling and methods and at least one community health worker (CHW) per village who provides education and counseling in community gathering places and home visits for women with babies and small children. CHWs can provide injectables, pills, condoms, and emergency contraceptives, and refer women for further services such as implants, IUDs, and sterilization. Contraception services are also available through the private sector in private and non-governmental clinics and pharmacies, though these outlets may be in accessible to many women in rural areas due to financial constraints.

This arm will receive the I-CAN intervention

I-CAN leverages four social support mechanisms (informational, instrumental, appraisal, emotional) to improve women's agency related to contraceptive decisions and actions. Community-based organizations (CBOs) recruit and train lay women ("mentors") via an intensive four-day training with a mix of didactic and practical sessions. Trained peer mentors will provide tailored social support to interested women in the community, drawing on their lived experiences to "mentor" other women in their village whom they recruit through community outreach. "Mentees" can opt-in to get tailored support (in the form of one-on-one consultations and/or accompaniment to health services) over a period of time they choose. Mentors are trained to provide neutral support and respect women's decisions

This is defined as the reported current use or non-use of any contraceptive method that aligns with one's stated preferences for using or not using any method. Participants will be asked 1) whether they or their partner is currently doing anything or using any method to delay or prevent pregnancy and 2) whether they want to do anything or use any method to delay or prevent pregnancy. A binary indicator will be constructed where 1=alignment between behaviour (current use or non-use) and preference (i.e., either using contraception and want to use contraception, or not using contraception and do not want to use contraception) and 0=misalignment (i.e., using contraception but do not want to or not using contraception but want to use contraception).

A 1-item measure using a 5-point Likert scale where 1=very dissatisfied and 5=very satisfied.

A 1-item measure with a binary response of 1=Has ever self-injected and 0=Has never self-injected.