Study setting {9}
The study will be conducted in Suba South Sub-County, Homa Bay County, Kenya. As of the 2019 national census, the sub-county had a population of 122,383 in 27,635 households, primarily composed of the Luo and Suba ethnic groups (8). The primary occupations are fishing and farming. Housing typically consists of multiple structures, with mud walls and metal sheet roofs being the most common, although iron sheets, concrete, or stone construction can also be found (8).
The Lake Victoria region generally experiences two rainy seasons annually: a long rainy season from March to June and a shorter one from August to October, although recent years have seen irregular patterns. Malaria incidence peaks one to two months after these rainy periods, with Anopheles (An.) gambiae s.s., An. arabiensis, and An. funestus being the primary vectors (9).
Suba South Sub-County is served by 30 public health facilities, including two Sub-county hospitals, eight health centers, and 20 dispensaries. The sub-county is divided into community health units, each monitored by a community health promoter (CHP) responsible for malaria case detection and treatment using RDTs and ACTs, respectively. The government periodically distributes LLINs free of charge, with the most recent distribution completed in 2023/2024.
Eligibility criteria {10}
Eligibility for the second-round intervention will be based on participation in the first-round experiment (2). All households that participated in the first-round baseline survey and provided consent for the follow-up surveys will be eligible for inclusion in the second round, regardless of their experimental arms in the original study. For detailed inclusion and exclusion criteria, please refer to our previous protocol (2) and Table 1 below.
Table 1
Inclusion and exclusion criteria for the intervention and prospective cohort survey
| Inclusion criteria | Exclusion criteria |
| First-round Baseline (survey) |
| At least one permanent resident aged 18 years or older in the household | |
| Informed consent provided by at least one adult in the household | |
| Baseline (blood sampling) |
| Members residing in the households participated in the baseline survey | Severe chronic illnesses |
| Informed consent provided by the parent or guardian before each survey | |
| Second-round financial incentive intervention |
| Household members residing in the households participated in the baseline survey | |
| Informed consent provided by at least one adult in the household | |
| Second-round follow-up 1 (survey) |
| Households participated in the baseline survey | |
| Informed consent provided by at least one adult in the household | |
| Second-round follow-up 1 (blood sampling) |
| Household members residing in the households participated in the baseline survey | Severe chronic illnesses |
| Informed consent provided by the parent or guardian before each survey | |
| Second-round follow-up 2 (survey) |
| Households participated in the baseline survey | |
| Informed consent provided by at least one adult in the household | |
| Second-round follow-up 2 (blood sampling) |
| Household members residing in the households participated in the baseline survey | Severe chronic illnesses |
| Informed consent provided by the parent or guardian before each survey | |
Who will take informed consent? {26a}
Written informed consent will be obtained by trained study team members who are fluent in Luo (a local language), Swahili, and English, and have a thorough understanding of the study protocol. After confirming eligibility, these team members will present potential participants with a comprehensive information sheet detailing the study’s purpose and contents as well as their right to withdraw at any time. The information sheet will be available in both Luo and English. For participants who are unable to read, the information will be conveyed verbally by a study team member. Participants will be given ample opportunity to ask questions and discuss any concerns before being asked to provide their consent. Written consent will be obtained only after the participant demonstrates a clear understanding of the study and voluntarily agrees to participate.
Additional consent provisions for collection and use of participant data and biological specimens {26b}
In addition to the study overview, the study information sheet will provide detailed information regarding the collection, storage, and use of personal data and biological specimens obtained during baseline and follow-up surveys, as well as blood sampling procedures.
Interventions
Explanation for the choice of comparators {6b}
LLINs are the most widely used malaria preventive measure in Kenya. The Kenya National Malaria Control Programme coordinates free LLIN distribution every three years, and county governments are responsible for delivery to residents in endemic areas. As all target households in this trial will have unrestricted access to LLINs through this program, the control group will reflect the current standard of prevention and care in the region.
The primary purpose of this trial is to evaluate the incremental effect of financial incentives combined with malaria education on top of existing LLIN coverage and usual care practices. Therefore, the control group households are expected to engage in standard malaria prevention and treatment-seeking behaviors, including LLIN use, as dictated by the national program and individual choices.
Intervention description {11a}
Our survey teams will visit households in all target clusters. For households in the control clusters, the teams will explain the upcoming follow-up survey at three and six months, confirm continued study participation, and conduct a short interview. In the intervention clusters, the team will also explain the follow-up surveys and confirm participation, but additionally inform households that they are part of a pilot intervention. They will then explain that eligibility for the intervention will be determined by a scratch card draw, with numbers 1–8 qualifying for the intervention (malaria education and a financial incentive) and numbers 9–10 not. The survey team will ensure the household representative understands the incentive scheme before the draw and will re-explain if necessary. For households selected for the intervention through the scratch card draw, the survey team will explain the specific reward scheme and the timing of the follow-up surveys at three and six months, and their members are invited to watch the malaria education material on a tablet device and then proceed to a short interview. Households unselected for the intervention will be informed that they will not receive the intervention and proceed to a short interview only. The detailed instruction script for enumerators throughout this intervention process is provided in Appendices 1 and 2.
Criteria for discontinuing or modifying allocated interventions {11b}
Given the low-risk nature of the financial incentive intervention and the absence of anticipated direct health risks, discontinuation of the intervention for individual participants will primarily occur under two circumstances. They include participant request (when a participant explicitly requests to withdraw from the study) and loss to follow-up (when a participant is lost to follow-up, including migration). Crossover from the control arm to the intervention arm will not be permitted during the follow-up period. Participants who migrate between study arms or emigrate from the study areas will be dropped from the study follow-up.
Strategies to improve adherence to interventions {11c}
In this study, adherence refers to the continued participation of individuals from both intervention-eligible and intervention-ineligible households throughout the follow-up period. Adherence will be monitored during the follow-up surveys by confirming the presence of participants and collecting the required blood samples.
Relevant concomitant care permitted or prohibited during the trial {11d}
There are no restrictions on concomitant care during the trial. All study participants, regardless of their assigned arm, are permitted to continue their usual healthcare practices, including the use of freely distributed LLINs and access to standard malaria testing and care through both public and private healthcare providers.
Provisions for post-trial care {30}
All study participants will remain under the standard care of the existing health system in Homa Bay County throughout and after the trial. Given the low-risk nature of the intervention, no specific post-trial care is anticipated to be necessary beyond the usual medical services available in the community.
Outcomes {12}
The primary outcomes of the study are (i) malaria prevalence by PCR in children and adults at three and six months post-intervention, (ii) change in the use of LLIN at three and six months post-intervention, and (3) change in malaria knowledge and perception among community residents at three and six months post-intervention. The secondary outcome is malaria prevalence by RDT in children and adults at three and six months post-intervention.
Participant timeline {13}
This study is the second round of a cRCT evaluating the impact of educational and financial incentives on malaria prevention and treatment behaviors (2). The first round of the study included the recruitment of participants and the collection of baseline data. As such, no additional recruitment is planned for this second round. Instead, the existing cohort of participants will be re-randomized into intervention and control arms to assess the effectiveness of a refined intervention. This sequential approach allows us to build upon the existing data and infrastructure, maximizing efficiency and minimizing disruption to the community.
Table 2
| | STUDY PERIOD |
| | Enrolment | Allocation | Post-allocation | Close-out |
| TIMEPOINT (month) | -19 | 0 | 3 | 6 | 18 |
| ENROLMENT: | | | | | |
| Eligibility screen | X | | | | |
| Informed consent | X | | | | |
| Allocation | | X | | | |
| INTERVENTIONS: | | | | | |
| Education + Incentivization (CCT) | | X | | | |
| Education + Incentivization (LIS) | | X | | | |
| ASSESSMENTS: | | | | | |
| Questionnaire survey | X | X | X | X | X |
| Capillary blood sampling by lab tech | X | | X | X | |
Sample size {14}
Based on an independent survey conducted concurrently in our study area, the estimated malaria RDT positivity rate is approximately 30%. With an average household size of 5 (as per our census data), we aim to detect a combined direct and spillover effect of our intervention, resulting in a 19.6% (or 5.89 percentage-point, ppt) reduction in RDT positivity for directly treated individuals and a 21.0% (or 6.30 ppt) reduction for untreated individuals within intervention clusters.
We retain the same cluster definitions and sample size as the first-round intervention (2), totalling 9,200 individuals from 1,840 households grouped into 92 clusters. With no results from the previous study yet available, the sample size and treatment assignment probabilities are calculated ex-ante. To achieve sufficient statistical power to detect the aimed effect sizes, we require a 27.32% RDT positivity reduction (or 8.20 ppt) for directly treated individuals and a 31.49% reduction (or 9.45 ppt) for untreated individuals within intervention clusters. Our calculation based on the optimal saturation design (7) assumes a power of 0.8, a two-sided type-I error of 5%, a cluster size of 20 households, and an intra-cluster correlation coefficient of 0.05. This calculation results in 28 clusters randomly assigned to the control condition and 32 clusters to each intervention condition (CCT and LIS). Within each intervention cluster,16 households are expected to receive the treatment, while the remaining four will serve as untreated controls. The two intervention arms will be compared to the control arm to assess their relative effectiveness.
Recruitment {15}
No new participants will be recruited for this second-round intervention. All households participating in the first-round study will be included in the second round, provided that they meet the continued eligibility criteria. For details on the initial recruitment, community sensitization, and census procedures, please refer to our previous protocol (2).
Assignment of interventions: allocation
Sequence generation {16a}
A new set of random numbers, independent of those generated in the first round, will be generated in the for each cluster using the statistical software R. These numbers will follow a uniform distribution between 0 and 1 with the seed set to the date of randomization in ‘yyyymmdd’ format (e.g., 20240321 for March 21st, 2024). R’s built-in functions set.seed() and runif() will be used for this procedure.
Concealment mechanism {16b}
Following stratification based on the malaria prevalence among children under 15 and the first-round intervention arm, a random number will be assigned to each cluster. The stratification creates six strata (2 x 3). Within each stratum, clusters will be sorted in ascending order based on their assigned random numbers. The clusters with the lowest random numbers (four to five per stratum) will be assigned to the control arm, followed by the CCT arm (five to six per stratum), and finally the LIS arm (five to six per stratum). This will result in a total of 28 control clusters, 32 CCT clusters, and 32 LIS clusters.
Implementation {16c}
A designated team member will generate the cluster-level allocation sequence on the day the intervention begins. Local study assistants will enroll participants and implement the assigned interventions in the respective clusters.
Assignment of interventions: Blinding
Who will be blinded {17a}
Cluster-level arm assignment will not be disclosed to trial participants. However, due to social interactions, participants may become aware of different arm assignments across clusters. We will assess the extent of this information flow through surveys.
Within intervention arms (CCT and LIS), household representatives will be aware of their household’s treatment status due to the lottery scratch card randomization.
Study team members participating in field activities cannot be blinded due to their involvement in cluster and household randomization. However, laboratory- and office-based personnel (e.g., microscopists and laboratory technicians) will remain blinded to participant identity and intervention status, as biological specimens will be identified by unique numeric study identifiers, and personal information will be removed before analyses.
Procedure for unblinding if needed {17b}
As this is a low-risk socio-economic intervention with no anticipated physical risks to participants, the need for unblinding procedures is currently not foreseen.
Data collection and management
Plans for assessment and collection of outcomes {18a}
Malaria prevalence will be estimated using cross-sectional malariometric surveys. The baseline survey was conducted at the beginning of the first-round study, and two follow-up surveys for the second-round experiment will be conducted approximately three and six months post-intervention. Malaria status will be determined using RDT, microscopy, and PCR, following the same procedures as described in the first-round protocol (2).
Malaria incidence will be assessed at both three- and six-month follow-up surveys, using the same procedure as described in the first-round protocol (2). During household visits, enumerators will administer a structured questionnaire to each participating household, collecting data on any history of fever, malaria episodes, or visits to local health facilities within the previous three months (10).
Enumerators and certified medical laboratory staff will be trained on the questionnaire's content and the use of the CSPro application data collection, adhering to the same procedures outlined in our first protocol (2). This includes built-in validation and completion checks to ensure data quality and completeness. To avoid duplication, all sample collection materials, such as microscope slides, filter papers, and sample tubes, will be pre-labeled with auto-generated serial numbers. The CSPro application will prompt staff to confirm participant identity and serial number before blood sampling, and the completeness of sampling will be verified twice: in the field and at the laboratory.
Plans to promote participant retention and complete follow-up {18b}
We will employ the same participant retention and follow-up strategies that proved successful in our first-round intervention (2). Specifically, our survey team will schedule household visits in advance, confirming the date and time with participants at least one week prior. CHPs will report any participant concerns or issues to the research team, and collaborative discussions will be held to address these issues promptly. Additionally, the research team will periodically accompany enumerators during household visits to reinforce the importance of the study and maintain participant engagement.
To enhance participant retention, we will implement a multi-faceted approach incorporating culturally tailored strategies, reflecting our established relationship with the community. This will include regular SMS and phone call reminders to participants, as well as personalized home visits by CHPs. These CHPs will serve as a trusted point of contact, addressing any questions or concerns participants may have and reinforcing the importance of their continued participation.
Data management {19}
Data management for this second-round intervention will largely follow the established procedures from the first round, as detailed in our previous protocol (2). Data will be collected on Android-based tablets using the CSPro application, incorporating built-in data validation checks. Follow-up survey data will be uploaded to the CSPro server, and after quality verification by data managers, local data will be deleted from the tablets to prevent overwriting. Each visit will be pre-programmed within CSpro for streamlined data entry. Specially trained enumerators who are locally hired and familiar with the participants will conduct the interviews, verifying participant identity before data entry. Data managers will perform periodic data quality checks on the server. Access to the survey data will be restricted to the study’s data analysts and managers.
Confidentiality {27}
To maintain confidentiality, each study participant is assigned a unique identifier. Personally identifiable information unnecessary for data analysis will be removed. The anonymized data will be stored separately from the key linking personal information on a secure server accessible only to authorized research staff. While publications will primarily contain aggregated data, anonymized individual-level data may also be included to enhance transparency and reproducibility, ensuring that no personally identifiable information can be derived from the published information.
Plans for collection, laboratory evaluation and storage of biological specimens for genetic or molecular analysis in this trial/future use {33}
Blood samples will be collected to examine malaria infection status using multiple methods, characterize immune responses to malaria parasites and mosquito saliva, and conduct malaria parasite genomic analyses. No human genetic studies are planned for this study. However, any remaining biological specimens after these analyses will be stored indefinitely for future research purposes unless the participants explicitly opt out during the informed consent process. Participants are provided with the study team’s contact information in the consent form and may withdraw from this study or any future research using their samples at any time without penalty or loss of benefits.
Statistical methods
Statistical methods for primary and secondary outcomes {20a}
The intention-to-treat (ITT) analysis will be the primary approach for both the primary and secondary outcomes. The impact of the intervention on these outcomes will be assessed using the regression framework as proposed by the prior study for a two-stage cRCT (7). This analysis will allow us to estimate the ITT effects, both direct and spillover, on treated and untreated households within a cluster while controlling for potential confounders such as age, gender, house structure, and socioeconomic indicators. A similar regression approach will be employed for the analysis of secondary outcomes. As a supplementary analysis, we plan to conduct and a per-protocol analysis, which will include only those participants who adhere to the study protocol and complete all required follow-up assessments. This analysis will provide additional insights into the intervention’s effect under ideal conditions of full compliance.
In addition to the primary and secondary analyses for the second-round intervention, we plan to conduct a combined analysis of the data from both the first and second rounds. This analysis will leverage Bayesian updating techniques to integrate the findings from both studies and obtain a more precise and comprehensive estimate of the intervention.
Interim analyses {21b}
We do not plan an interim analysis.
Methods for additional analyses (e.g., subgroup analyses) {20b}
Given the known association between malaria risk and socioeconomic factors (11), we plan to conduct subgroup analyses to explore potential heterogeneity in the intervention’s effects across different socioeconomic backgrounds. These analyses will utilize relevant socioeconomic indicators collected during the baseline survey. The same regression model employed for the primary analysis (7) will be used to estimate treatment effects within each subgroup. If feasible, we intend to assess the interaction between the intervention and socioeconomic variables to formally test whether the intervention’s effect differs across subgroups defined by socioeconomic variables.
Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data {20c}
Once all data collection is completed, the extent and patterns of missing data will be assessed. If necessary, multiple imputation methods may be used to handle missing data.
Plans to give access to the full protocol, participant-level data and statistical code {31c}
This manuscript is the full protocol. The authors will create anonymized datasets. Future statistical codes will be made available upon reasonable request.
Oversight and monitoring
Composition of the coordinating centre and trial steering committee {5d}
A coordinating centre has been established to oversee the trial's day-to-day operations. This centre includes a communication group comprising enumerators and laboratory technicians, who exchange experiences and address operational challenges. A local management team, consisting of study investigators from Kenya and Japan, provides guidance and ensures sample and data integrity. A steering committee, composed of key researchers from Kenya and Japan, including the principal investigator (PI) and co-PI, will convene monthly meetings to monitor overall trial progress and make strategic decisions.
Composition of the data monitoring committee, its role and reporting structure {21a}
Given the low-risk nature of the intervention, a formal data monitoring committee has not been established for this study. However, to ensure data quality and methodological rigor, the research team will consult with an independent statistician as needed for additional review and validation of statistical analyses.
Adverse event reporting and harms {22}
The financial incentives and malaria education are not expected to pose significant risks. However, should any adverse events occur during the study period, participants will be instructed to report to their designated CHP, who will then promptly inform the research team and Homa Bay County Ministry of Health (MOH). MOH’s medical officers will assess the relatedness of each reported event. In the event of a serious adverse event suspected to be study-related, the PI will be immediately notified, and the study team, in consultation with the MOH and Homa Bay County Teaching and Referral Hospital representatives, will convene an urgent meeting to review the case and determine its relatedness to the study. Appropriate action, such as protocol modifications, additional medical care, or study termination, will be taken if the case is deemed related to the study and considered to do excessively more harm than good to participants and communities.
Frequency and plans for auditing trial conduct {23}
Regular monitoring will occur throughout the study to ensure adherence to the protocol and data quality. This can include a pre-implementation review of all study procedures, monthly progress meetings during follow-up, annual reports and renewal applications to the relevant IRBs (Otaru University of Commerce, Japan, and Mount Kenya University, Kenya), and regular data quality checks.
Plans for communicating important protocol amendments to relevant parties (e.g., trial participants, ethical committees) {25}
Any substantial protocol modifications that may impact the study’s conduct, potential benefits or harm to participants, or participant safety will be submitted to the relevant institutional review boards (IRBs) at Mount Kenya University and Otaru University of Commerce for approval prior to implementation. Upon approval, such amendments will be communicated to study participants and updated in the trial registry. Minor amendments not impacting participant safety or study validity will be documented and reported to the IRBs.
Dissemination plans {31a}
The results of this study will be shared with relevant stakeholders, including the Homa Bay County Government and Kenya National Malaria Control Program, to discuss the potential for scaling up the intervention. Findings will also be disseminated through peer-reviewed publications and presentations at scientific conferences, with the aim of informing the development of novel malaria control strategies in other malaria-endemic regions. Furthermore, feedback from research participants will be actively sought and incorporated into future refinements of the intervention, ensuring its continued relevance and acceptability within the community.