Integrating nutrition communication in agricultural intervention programs aimed at increased food availability and accessibility in resource-poor areas is crucial. To enhance the sustainability and scalability of nutrition communication, the present study piloted the approach of ‘nutrition integrated agricultural extension’ and tested nutrition-related outcomes with two types of nutrition messages (specific vs. sensitive) and two delivery channels (public sector vs. private sector). The study intervention comprised (i) vegetable seed kit distribution, (ii) ongoing agricultural extension activities by public or private sectors and (iii) nutrition communication with two different messages. The intervention was tested with three treatment arms and reached 454 farmers (>65% female) in rural Kakamega County, Western Kenya. Pre-/post-surveys measured outcome variables focused on farmers’ nutrition-related knowledge, attitudes and practices in vegetable production and consumption, and household dietary diversity score. Results showed that all treatments increased nutrition knowledge (p < 0.05). Nutrition-specific communication was more effective than nutrition-sensitive communication. Nutrition communication through either the public or the private agricultural sector was both effective. Before the study intervention, many participants believed that vegetable consumption was beneficial and wanted to increase intake. After the intervention, the number of participants who felt eating more vegetables was challenging decreased slightly. Nutrition communication was found to be especially important in conveying recommended food amounts and promoting increased vegetable consumption. Seasonality affected on-farm crop diversity and vegetable consumption results in this study.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.