Background: Almost a quarter of the world’s undernourished people live in India. We tested the effects of three nutrition-sensitive agriculture (NSA) interventions on maternal and child nutrition in India.

Methods: We did a parallel, four-arm, observer-blind, cluster-randomised trial in Keonjhar district, Odisha, India. A cluster was one or more villages with a combined minimum population of 800 residents. The clusters were allocated 1:1:1:1 to a control group or an intervention group of fortnightly women’s groups meetings and household visits over 32 months using: NSA videos (AGRI group); NSA and nutrition-specific videos (AGRI-NUT group); or NSA videos and a nutrition-specific participatory learning and action (PLA) cycle meetings and videos (AGRI-NUT+PLA group). Primary outcomes were the proportion of children aged 6–23 months consuming at least four of seven food groups the previous day and mean maternal body-mass index (BMI). Secondary outcomes were proportion of mothers consuming at least five of ten food groups and child wasting (proportion of children with weight-for-height Z score SD <–2). Outcomes were assessed in children and mothers through cross-sectional surveys at baseline and at endline, 36 months later. Analyses were by intention to treat. Participants and intervention facilitators were not blinded to allocation; the research team were. This trial is registered at ISRCTN, ISRCTN65922679.

Findings: 148 of 162 clusters assessed for eligibility were enrolled and randomly allocated to trial groups (37 clusters per group). Baseline surveys took place from Nov 24, 2016, to Jan 24, 2017; clusters were randomised from December, 2016, to January, 2017; and interventions were implemented from March 20, 2017, to Oct 31, 2019, and endline surveys done from Nov 19, 2019, to Jan 12, 2020, in an average of 32 households per cluster. All clusters were included in the analyses. There was an increase in the proportion of children consuming at least four of seven food groups in the AGRI-NUT (adjusted relative risk [RR] 1·19, 95% CI 1·03 to 1·37, p=0·02) and AGRI-NUT+PLA (1·27, 1·11 to 1·46, p=0·001) groups, but not AGRI (1·06, 0·91 to 1·23, p=0·44), compared with the control group. We found no effects on mean maternal BMI (adjusted mean differences vs control, AGRI –0·05, –0·34 to 0·24; AGRI-NUT 0·04, –0·26 to 0·33; AGRI-NUT+PLA –0·03, –0·3 to 0·23). An increase in the proportion of mothers consuming at least five of ten food groups was seen in the AGRI (adjusted RR 1·21, 1·01 to 1·45) and AGRI-NUT+PLA (1·30, 1·10 to 1·53) groups compared with the control group, but not in AGRI-NUT (1·16, 0·98 to 1·38). We found no effects on child wasting (adjusted RR vs control, AGRI 0·95, 0·73 to 1·24; AGRI-NUT 0·96, 0·72 to 1·29; AGRI-NUT+PLA 0·96, 0·73 to 1·26).

Interpretation: Women’s groups using combinations of NSA videos, nutrition-specific videos, and PLA cycle meetings improved maternal and child diet quality in rural Odisha, India. These components have been implemented separately in several low-income settings; effects could be increased by scaling up together.