Increasing household diet diversity and food security in rural Rwanda using small-scale nutrition-sensitive agriculture: A community-level proof of concept study

Objectives In Rwanda, rates of malnutrition remain high in rural areas where residents consume a primarily starch-based, low variety diet. Nutrition-sensitive agricultural interventions using kitchen gardens have been effective in addressing low diet diversity in similar populations. This studys objective was to develop a kitchen garden and nutrition education intervention aimed at sustainably increasing diet diversity and food security at the household-level. Design A mixed methods community-level study, with a sixteen-week nutrition-sensitive agricultural intervention including nutrition education was conducted. Household diet diversity scores and household hunger scores were calculated at baseline, post-intervention and one-year follow-up. Setting The intervention was conducted in a rural Rwandan community in the Northern Province. Participants Stratified purposeful sampling techniques were used to select women participants representing forty-two households. Results Household diet diversity scores increased over time from pre-intervention to six months post-intervention and one-year post-intervention. The magnitude of the change was similar in all stratified groups (2.3x at 6 months and 2.9x at 1 year). Households whose main source of income was working for other farmers, reported a significantly lower diet diversity score than those households receiving income from sources [t(40) = -2.108, p=0.041]. Among those households not consuming protein and vitamin-A rich food groups at baseline, all reported consuming foods from these food groups post-intervention. There were no significant changes in household hunger scores. Conclusions Collaborative community-based nutrition-sensitive agricultural interventions using kitchen gardens, can increase household diet diversity, which may encourage sustained change in dietary patterns for nutritional adequacy in low-income rural Rwandan populations.

This method was appropriate because of the homogeneity of the groups and the research goal of 134 examining variation in key variables. 16 Randomized sampling was infeasible as the long-term 135 goal of the intervention was to provide sustained change and generate information spread within 136 a specific community. Therefore, by allowing community liaisons, who were respected members 137 of the community, to recruit and choose participants based upon the study criteria, it allowed for 138 greater community buy-in, which was considered essential to the success of the intervention. 139 Subjects were recruited from each of the six, government-established geographically specific 140 community groups, referred to as cells. Each cell contained seven participants, and a leader for 141 each cell was chosen internally by the group. The initial forty-two participants were referred to 142 as project ambassadors to foster empowerment and emphasize their future role as trainers. This . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022

Week
Learning Topic 1 Introduction to basic nutrition and home gardens 2 Trench garden design 3 Keyhole garden design 4 Work week 5 Water management and the nutritional needs of various household members part 1 6 Composting and different nutrients in food 7 Work week 8 Pest and weed management and the nutritional needs of various household members part 2 9 Seed saving and household meal planning 10 Work week 11 Perennial crops, fruit trees and nutritional deficiencies 12 Food safety and preservation and ensuring household food security 13 Cooking and planning balanced meals 14 Work week 15 Question and answer session 16 Large group reflection and celebration . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 Participants created keyhole-style and raised-bed gardens at their homes, as well as 159 compost piles and rainwater catchment systems for irrigation. They also learned about how to 160 prepare balanced meals for all members of their household using the fruits and vegetables they 161 grew in their kitchen garden. Resources were provided throughout the intervention, including 162 workbooks and notepads, gardening tools, seeds, and construction materials -Some resources 163 were intended for personal-use and others to be shared. Due to the varied education and literacy 164 levels, materials included primarily pictures and illustrations. Semi-structured interviews were used to collect data on diet diversity and food security. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 research team determined the interviewer and translator to be able to collect accurate data in 181 accordance with the study protocol.  food groups consumed by individuals in a household over the 24-hour recall period. 9 Information 200 recorded from the 24-hour recalls was used to determine a household diet diversity score 201 (HDDS) based upon adaptation of the Food and Agriculture Organization (FAO) tool to include 202 regional and cultural-specific foods. 19 For the purpose of discussion, protein foods are defined as . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ; https://doi.org/10.1101/2022.09.27.22280437 doi: medRxiv preprint foods from groups: Organ and Flesh Meat,Eggs,Fish and Seafood,Legumes,Nuts and Seeds,204 and Milk and Milk Products. HDDS is a qualitative measure that can serve as a proxy for the 205 nutrient adequacy of the diet of individuals while also reflecting household access to a diversity 206 of foods. 9,19,20 The HDDS is calculated based on the number of food groups consumed out of 207 twelve possible food groups ( Table 2). 19   tests. For all statistical analysis, a P value of ≤ 0.05 was accepted as significant.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint  When divided into cell groups, the average HDDS continued to increase amongst most 279 participant households in most cell groups (Figure 3.). This sustained increased in HDDS was . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

(which was not certified by peer review) preprint
The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 also seen when HDDS was analyzed for each cell group over time. There was a statistically  To determine if the changes in household diet diversity contributed to improved 298 nutritional adequacy, the proportion of participants consuming vitamin A-rich foods and protein 299 foods by cell group at the three time points was examined (Figure 5a. and Figure 5b.). Among 300 those participants not consuming protein foods and vitamin A-rich vegetables at baseline, after 301 the intervention, all reported consuming foods from these food groups (as defined in Table 2.). In 302 addition, a large majority (83%) of the participants growing vitamin A-rich vegetables in their 303 kitchen gardens, also consumed these vegetables. Thirty-one percent of participants who were 304 consuming 'other' vegetables, were also growing them. However, 42% of participants who were 305 growing 'other' vegetables, were not consuming them. The vegetables grown at the highest rate . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 were amaranth leaves, a dark green leafy vegetable known locally as dodo (57%), onions (55%), 307 green cabbage (52%), beets (50%), and carrots (43%). The results for household food security are represented in Figure 6. and 7. There were no 314 significant changes in HHS scores for any participants across all timepoints, or when divided 315 into cell groups. There were also no significant differences in HHS when examining scores 316 amongst cell groups. Additionally, no relationship was found between changes in HDDS and 317 observed changes in HHS. It was observed however, that households whose main source of  It was hypothesized that the nutrition-sensitive kitchen garden intervention would result 332 in a sustained increase in household diet diversity and food security, allowing for improvement 333 in dietary patterns aimed at reducing malnutrition. It was observed that the average household . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 diet diversity increased six-months after the intervention, during the sunny season, as well as increases in all food groups except a few (Figure 1.). Unlike diet diversity measurements for 342 individuals, when measuring household diet diversity there is no standardized cutoff of food 343 groups that will equal nutritional adequacy, thus a discussion of patterns is better suited for this 344 data. 25 At first glance, it appears the biggest increase is observed in the Spices, Etc. food group 345 with an increase in over 80%, which could be interpreted as inconsequential as the gardens did 346 not provide salt directly. However, upon further analysis, iodized table salt was the only spice 347 reported, providing the essential micronutrient iodine, often lacking in the diets of poor rural 348 populations globally. 26 In addition, the consumption of many nutrient-dense foods increased by 349 at least 20% from baseline to September 2019 to include foods in the following groups: Cereals, 350 Vegetables, Oils & Fats, Fruits, and Legumes, Nuts & Seeds. Within the Vegetables food group, 351 variety and nutrition density increased substantially, particularly with participants consuming 352 more colorful varieties such as spinach, beets, carrots, and onions, providing health-promoting 353 phytochemicals. They also increased the consumption of Vitamin A-rich vegetables such as dark 354 leafy greens and carrots. However, since no biomarker data was collected on actual 355 micronutrient levels in participants, no conclusions can be reported on actual nutrient status of 356 participants or household members -a limitation that could be directed toward future study.
. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ; https://doi.org/10.1101/2022.09.27.22280437 doi: medRxiv preprint cases milk and fish. Thus, it is possible that kitchen gardens not only increased the consumption 368 of a variety of nutrient-dense vegetables, but also allowed for foods in other food groups to be 369 consumed due to more income flexibility. Increasing income flexibility in a household is an 370 indicator of progress toward poverty reduction by allowing for the purchase of nutrient-dense 371 foods, rather than just those that meet energy needs. 6,27 Therefore, further evaluation is warranted 372 of the impacts of this intervention approach on income flexibility as a future research direction.

373
These changes in diet diversity did vary among the different cell groups. The reason is 374 unclear but could be related to differences in the income source of the household among the 375 groups. As reported, households whose main source of income was from working on the farms 376 of others, saw lower diet diversity then those households that had someone working in the 377 market or having formal employment (Figure 4.). This has also been reported in similar rural 378 populations where those having a higher, more stable income source was associated with greater 379 diet diversity. 6 It is possible that the informal economy observed in rural areas where agriculture . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022 study in Kenya and Uganda, reported decreases in diet quality and increased food insecurity The collaborative structure of the study within the community could result in response 444 bias, as participants are also members of the community, and thus have a stake in the outcomes.

445
This is especially true when the community and/or participants are receiving resources. In an 446 attempt to mitigate this risk, triangulated data collection methods and participatory action 447 research evaluation methods were employed. Second, we used household diet diversity scores as . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted September 30, 2022. ;https://doi.org/10.1101https://doi.org/10. /2022