Small mobile conditional cash transfers (mCCTs) of different amounts, schedules and design to improve routine childhood immunization coverage and timeliness of children aged 0-23 months in Pakistan: An open label multi-arm randomized controlled trial

Background: Cost-effective demand-side interventions are needed to increase childhood immunization. Multiple studies find tying income support programs (≥USD 50 per year) to immunization raises coverage. Research on maximizing impact from small mobile-based conditional cash transfers (mCCTs) (≤USD 15 per fully immunized child) delivered in lower-income settings remains sparse. Methods: Participants in Karachi, Pakistan, were individually randomized into a seven arm, factorial open label study with five mCCT arms, one reminder (SMS) only arm, and one control arm. The mCCT arms varied by amount (high ∼USD 15 per fully immunized child versus low ∼USD 5 per fully immunized child), schedule (flat versus rising payments over the schedule), design (certain versus lottery payments), and payment method (airtime or mobile money). Children were enrolled at BCG, pentavalent-1 (penta-1) or pentavalent-2 (penta-2) vaccination and followed until at least 18 months of age. A serosurvey in 15% sub-sample validated reported study coverage. The full immunization coverage (FIC) at 12 months (primary outcome) was analyzed using logit regression. ClinicalTrials.gov (NCT03355989), 3ie registry (58f6ee7725fc1), and AEA RCT Registry (AEARCTR-0001953). Findings: Between November 6, 2017, and October 10, 2018, a total of 11,197 caregiver-child pairs were enrolled, with 1598-1600 caregiver-child pairs per arm. FIC at 12 months was statistically significantly higher for any mCCT versus SMS (OR:1.18, 95% CI: 1.05-1.33; p = 0.005). Within the mCCT arms, FIC was statistically significantly higher for high versus low amount (OR: 1.16, 95% CI: 1.04-1.29; p = 0.007), certain versus lottery payment (OR: 1.30, 95% CI: 1.17-1.45; p < 0.001) and airtime versus mobile money (OR: 1.17, 95% CI:1.01-1.36; p = 0.043). There was no statistically significant difference between a flat and increasing schedule (OR: 1.03, 95% CI: 0.93-1.15; p = 0.550). SMS had a marginally statistically significant impact on FIC versus control (OR: 1.16, 95% CI: 1.00-1.35; p ...