This study will focus on a new community health worker (CHW) program called 'Flourish', which is implemented by Philippine non-governmental organization (NGO) International Care Ministries (ICM). ICM's core program is 'Transform', which is a poverty alleviation program that focuses on households living in ultra-poverty (less than 0.50 United States dollars (USD) per person per day). Transform leverages local community networks to identify 30 participants from the most marginalized households in the community. ICM then provides these participants with 15 weeks of health and livelihood education, as well as with health interventions, including childhood malnutrition treatment.

In ICM's Flourish program, one of the 30 participants in each Transform program is nominated by their peers to become an ICM-supported CHW. The CHWs will not only provide health services after the conclusion of the Transform program for their fellow Transform participants' households, but they will also serve an additional 50+ households in their communities. These additional 'non-Transform' households will be identified by asking the Transform participants to nominate peers in their social networks that share similar socioeconomic backgrounds and whose households are perceived to also benefit from CHW services.

The primary aim of this research is to assess the impact of CHWs on the prevalence of acute childhood malnutrition in hard-to-reach households within low-income communities.

The investigators hypothesize that this study will show that CHWs are able to reduce the prevalence of acute childhood malnutrition in hard-to-reach households within low-income communities.

This study will focus on a new community health worker (CHW) program called 'Flourish', which is implemented by Philippine NGO International Care Ministries (ICM). ICM's core program is 'Transform', which is a 15-week poverty alleviation program that focuses on households living in ultra-poverty (less than 0.50 United States dollars (USD) per person per day). ICM runs 'Transform' in approximately 1,000 communities every year. Transform leverages local community networks to identify 30 participants from the most marginalized households in the community. ICM then provides these participants with 15 weeks of health and livelihood education, as well as with health interventions, including childhood malnutrition treatment. During the program, participants are given the opportunity to form a community savings group with the support of ICM. On average, twenty participants per community join these savings groups.

Participants receive a continuation of services from ICM after the Transform program ends. ICM promotes continued economic benefits by supporting the newly formed savings group. ICM also promotes continued health benefits through the 'Flourish' program. In the Flourish program, the savings group elects one person from their membership to serve as a community health worker. This community health worker is trained to screen for illness, provide health counseling, link to primary care, and deliver health commodities to: 1) the savings group's households and 2) the wider community. ICM equips each community health worker with a mobile phone and health commodities. The mobile phone assists the community health worker with screening and follow-up procedures and is also used to collect data on the household members.

A cluster randomized controlled trial will be run to assess the impact of the Flourish program on reducing malnutrition in children between 6 months and 12 years of age. A total of 170 communities will be randomized into two even groups: one arm will receive both the Transform and Flourish programs (Treatment) and one arm will receive the Transform program only (Control).

The maximum number of clusters that ICM can support during the study period is 170, therefore sample size calculations were focused on detectable effect size based on the number of children enrolled per cluster. With significance at 5% and power at 80%, given a background malnutrition rate of 8.5% (based on rates observed in similar settings by ICM) the investigators estimate the ability to detect a reduction of 3% points (35% decrease in malnutrition/ achieving 65% cure rate) if 40 children are enrolled per cluster. To account for potential attrition of ~10%, the study will target 45 children per cluster, which will be a cumulative target of 7,650 (45 x 170) for the whole study. The R package "clusterPower" was used for calculations.

In both arms, each member of the savings group will be asked to nominate up to three households to potentially be approached by ICM community health workers. The nominators will be asked to identify households that are of similar economic status and will be given criteria to help their assessment. Additionally, the households should have at least one child between the ages of 6 months and 12 years who has not previously received an ICM intervention. An ICM staff member will also approach the closest local primary health unit (Barangay Health Station or Rural Health Unit) and request nominations of any households with currently malnourished children.

The list of savings group households and their nominated households will be collected by an ICM staff member. Each household will receive an explanation that by enrolling in this study, they may be approached by a future ICM community health worker program. Only people enrolled in the Treatment arm of this study will be approached to participate in the community health worker program; however, all study participants will receive a free child malnutrition screening in February-March 2025. At that screening, any child between 6m-12 years who is identified as malnourished will be linked to care and provided treatment for malnutrition. Only consenting households will be enrolled into the study. In both arms, participation is completely voluntary, and participants are allowed to withdraw at any time and still receive the screening in 2025.

All households enrolled in the study will also receive a short baseline survey in July 2023 and endline survey in February-March 2025. Surveys will be conducted by trained enumerators. The baseline survey will primarily collect household identification and demographic information. The baseline data will also be analyzed to determine the balance between Treatment and Control participants on key characteristics.

In the Treatment arm, the community health worker will attempt to register all consenting savings group households into the Flourish program. They will also attempt to register the 50 additional households from the nominee list into the Flourish program. The CHW must obtain consent from the heads of households of their participation in the community health worker program before registering the household.

Over time, the community health worker may continue to enroll additional households in the Flourish program, including households who were nominated by the Savings Group but not originally registered in the Flourish program. Households who did not consent to the study may also be registered into the Flourish program, as long as they provide consent for registration. However, households who were not on the original list of nominees will not be included as study participants.

In February 2025, 18-months after household enrollment, enumerators will return to every household in both the Treatment and Control arms. The enumerators will conduct an endline survey and collect the age, weight, height, and mid-upper arm circumference of each child. In the Treatment arm, children identified as wasted who are not already on malnutrition follow-up will be flagged for the community health worker to register into the Flourish program (if not already) and receive malnutrition follow-up and treatment from the CHW. In the Control arm, children identified as wasted will be referred to the closest primary health service provider (Barangay Health Station or Rural Health Unit), and if provisions are not adequate ICM will follow up with these children and their households.

The following analyses will be conducted for the primary research objective to assess the impact of the Flourish program on malnutrition in children aged 6 months to 12 years old at time of enrollment:

• A logistic regression model with clustered standard errors will be used to analyze the difference in prevalence of malnutrition (WHZ <-2) between Control and Treatment.
• A linear regression model with clustered standard errors will be used to analyze the difference in mean weight-for-height Z-score between Control and Treatment.

Inclusion Criteria:

Participants in the study must meet at least one of the criteria:

• Transform program participant and member of the savings group formed at the end of Transform
• Nominated by a savings group member to participate in the study
• Nominated by the local public health unit (based on confirmed malnutrition status) to participate in the study

Exclusion Criteria:

• Except for members of the savings group, households cannot participate in the study if they have previously received ICM programs.