Bringing Greater Precision to Interactions Between Community Health Workers and Households to Improve Maternal and Newborn Health Outcomes in India

Estimated Relationships Between Health Behaviors and (A) Number of CHW Home Visits and (B) Timing of First Home Visit

Abbreviations: ANC, antenatal care; CHW, community health worker; IFA, iron and folic acid.Only significant associations are shown, see Table 1 for all associations. All means are estimated marginal proportions adjusted for covariates.^a n=5,438 women.^b n=4,541 women, excluding those who received no visit.

(A) Increase in Health Behavior if Household Member Reported Having Talked to CHW About the Particular Behavior^a; (B) Percentage of CHW Workforce That Has Ever Used Each Type of Message When Convincing Households to Attend Checkups^b; (C) Percentage of CHW Workforce That Has Ever Used Each Type of Message When Convincing Households to Deliver in Facility^c; (D) Predicting Whether a Woman Attended 3+ Checkups Based on the Preferred Behavior Change Message of the CHW^d; (E) Predicting Whether a Woman Had an ID Based on the Preferred Behavior Change Message of the CHW^e

Abbreviations: ANC, antenatal care; CHW, community health worker; ID, institutional delivery; IFA, iron and folic acid; MIL, mother-in-law; VHND, village health nutrition day.All rates represent adjusted means, including covariates for several indicators of CHW quality and performance. * Indicates P<.05 compared to the reference.^a Expressed as absolute increase with 95% confidence intervals (percentage points) based on regression. n=2,057 households that received CHW visit and had woman, husband, and MIL interviewed.^b Self-reported in the CHW survey. n=1,318 CHWs that had at least 1 household receiving a visit from them in the data.^c n=1,318 CHWs.^d The most commonly preferred message is used as reference, and differences represent absolute changes in the rate of behavior. n=4,134 women.^e N=4135 women.

(A) Relationship Between Outcomes and CHW Presence at Birth Irrespective of Location of Delivery^a; (B) Relationship Between Outcomes and CHW Presence at Birth by Location of Delivery^b; (C) Association of CHW’s Awareness of Delayed Bathing and Practice of Delayed Bathing When CHW Was Absent or Present^c; (D) Rate of Different Forms of Poor Treatment at Public Facility Deliveries When the CHW Was Absent or Present^d

Abbreviations: BF, breastfeeding; CHW, community health worker.^a N=5,240 women. * Indicates statistical significance at P<.05.^b N=984 women at home, 1068 at private, 3123 at public.^c N=4831 women.^d N=3244 women. Error bars represent 95% CI.