Tuberculosis (TB) remains the leading cause of death from a single infectious disease globally, with an estimated 10 million cases in 2018 ¹ . The WHO African region accounts for over 20% of TB cases, with Nigeria alone contributing 4% ¹ . Despite a decline in TB mortality in Nigeria ² , the incidence has stagnated since 2000, mainly due to delayed diagnosis and treatment ³ . Many TB-infected individuals are asymptomatic and unlikely to seek care, leading to significant underdiagnosis and posing a major challenge to TB control efforts ⁴ .

In 2018, 7 million new TB cases were reported globally, meeting the UN's target under the WHO initiative "Find. Treat. All. #EndTB" ¹ . However, an estimated gap of about 10 million cases persists between reported and estimated cases, with Nigeria, India, Indonesia, and the Philippines, accounting for over 50% of the underreported cases ¹ . This under-detection hampers national TB programs and highlights the limitations of traditional case-finding methods ^{3, 5} .

Active case-finding, recommended by the WHO, aims to bridge this gap by screening high-risk individuals, such as contacts of TB patients and people living with HIV. This method identifies and treats those with TB who would not otherwise seek healthcare, effectively detecting both active and latent TB ⁶ . Screening tools include history taking, sputum smears, and chest X-rays, enhancing early diagnosis and treatment ⁶ .

Nigeria has a low TB case detection rate of 25%, contributing to 12% of the global gap between TB incidence and reported cases in 2018 ⁷ . The South-South region of Nigeria reported 12% of the country's TB cases ⁸ . Furthermore, Akwa Ibom has the highest HIV prevalence at 5.5% ⁸ , which increases TB risk. There are pockets of high TB burden communities in some LGAs in the state, making Akwa Ibom one of the states with persistently low case detection of all forms of TB. Active case detection has been initiated in high-risk patient groups ⁴ . However, there are still gaps in active case-finding.

Many high-risk people such as co-infected, co-morbid, malnourished persons, and children with active TB do not experience typical TB symptoms in the early stages of the disease. They are therefore unlikely to seek care early and may not be properly diagnosed when they eventually seek care. Poor knowledge of symptoms and where to access care also affects the uptake of TB services in our communities.

Community health workers (CHWs) are crucial for providing primary health services in Nigeria, particularly in rural areas. There are different categories of CHWs, including Community Health Extension Workers (CHEWs), Community Pharmacists (CPs), and Patent Medicine Vendors (PMVs). CHEWs are formally trained and typically employed by the government, although many in Akwa Ibom state serve as volunteers ⁹ . Community Pharmacists offer medical services, dispense medications, and provide healthcare advice, often serving as the first point of contact due to their accessibility and shorter wait times ^{10– 13} . PMVs, on the other hand, are self-trained lay health workers who diagnose and treat various ailments with no formal qualifications, though the Pharmaceutical Society of Nigeria regulates their activities ¹⁴ .

Effective training of frontline health workers is essential for improving tuberculosis (TB) control. Studies have shown that enhanced knowledge and attitudes towards TB among health workers lead to better treatment and prevention outcomes. For instance, a training program in Mozambique found a 14.6% increase in TB case notification rate in the intervention group compared to a decrease of 16.5% in the control group ¹⁵ . A similar initiative in Enugu led to a doubling of TB presumptive cases and diagnoses three months after a health worker training was done ¹⁶ .

Incentives, particularly cash incentives, have been shown to boost the commitment, innovation, and accountability of health workers ^{17, 18} . While these incentives have proven effective in various health programs, there is limited evidence of their impact on improving TB case detection rates among community health workers in low- and middle-income countries.

This study aimed to assess the effectiveness of training and cash incentives for the improvement of TB case notification among CHWs in Akwa Ibom State, Nigeria. The main research questions were: Can cash incentives and training be used to motivate CHWs to increase TB case detection? The specific objectives were i) to determine the effect of cash incentives plus training on TB case detection ii) to determine the effect of training-only CHWs on TB case detection.

Akwa Ibom State is located in the South-South region of Nigeria, one of the states in the oil-rich Niger Delta region with a population of about 5.5 million based on 2006 National population census report ⁷ . The state has 31 local governments distributed across the three senatorial districts. It has 368 primary health centers (PHCs), unevenly distributed across the local government areas (LGAs) ¹⁹ .

Akwa Ibom state has a high burden of TB and HIV ⁸ . At the time of this study, the USAID was carrying out TB control activities in 15 LGAs and TB Reach had projects in three of the remaining 16 LGAs. The study population was the 13 LGAs that were not covered by the ongoing TB programs.

The interventions took place in the PHCs and catchment communities where the CHEWs, PMVs and CPs practised. While most of the CHEWs worked in the public sector PHCs, the CPs and PMVs worked in the private sector, setting up their drug stores with the sole aim of making a profit. The baseline interventions were carried out in April 2019 and follow-up was carried out quarterly until March 2020. The trial ended as scheduled after one year, although the COVID-19 pandemic lockdown interrupted endline data collection and close-out of the project.

The study was designed as a three-arm parallel cluster randomised control study. The interest in evaluating training and cash incentives concurrently informed the choice of a multi-arm cluster randomised trial. The advantages of this design include lowering costs as the two arms run at the same time and having one control arm which reduces the sample size relative to performing separate 2-arm trials ²⁰ .

It was conducted in six high-TB-burden Local Government Areas (LGAs). These LGAs were purposively selected from the sampling frame of 13 LGAs with the assistance of the State Tuberculosis and Leprosy Control Officer. Each local government area served as a cluster for this study, and they included Esit Eket and Uyo (Arm A), Ibiono Ibom and Ibesikpo (Arm B), and Nsit Ibom and Uruan (Arm C-Control).

The study clusters were randomised by the researchers using the balloting technique into one of the three arms. In each participating local government area, three PHCs offering DOTS treatment services were selected by simple random sampling and included in the study. In total, 18 PHCs and their catchment communities were used, with PHCs clusters per arm. The allocation of PHCs can be seen in Figure 1. Recruitment of CHWs in each arm was facilitated by the PHC focal persons, who worked with the Chairman of the Patent Medicine Vendors (PMVs).

Invitations were sent to all CHWs in the intervention PHCs (Arm A and B) and their catchment communities for a centrally conducted workshop in Uyo, the capital city. Blinding of participants to their allocated arms was not possible. To mitigate bias from non-blinding, clusters were allocated to the study arms by LGA, and selected LGAs were not contiguous. The workshop was conducted on two separate days: Arm A on Day 1, and Arm B on Day 2. Blinding assessors to the different arms was also not possible.

The PHCs served as the focal point and referral centre for any cases identified in the community by the CHWs.

1. Training of CHWs : A total of 158 HCWs were trained, 85 in Arm A and 73 in Arm B ( Figure 2) via a one-day training session conducted separately for each of the two intervention arms. This face-to-face workshop was facilitated by researchers in collaboration with the State Tuberculosis and Leprosy Control Program office.

Before the workshop, the research team and the Akwa Ibom State TB, Leprosy, and Buruli Ulcer Control Program (AKSTLBCP) State Coordinator conducted a “training of trainers’ session with workshop facilitators. The facilitators were trained on the course content to ensure a shared understanding of the workshop's aims and to familiarize them with the workshop materials.

Training manuals were developed specifically for the research and distributed to participants. Facilitators used both the training manuals and PowerPoint presentations developed from the manuals. Facilitators also used participatory learning methods to deliver the course contents. The contents of the training were based on the module developed for active TB case finding for community health workers through house-to-house search for community-based organizations (CBOs) and CHWs by the National Tuberculosis and Leprosy Control Program (NTBLCP) ²¹ . The sessions included training on basic symptoms, misconceptions diagnosis and treatment of TB, identification of presumptive TB cases, sputum collection and transportation and linking TB patients to care and treatment.

Participants were also taught how to collect and transport sputum samples by the State Laboratory Focal Person for TB, using sputum cups and transportation media (plastic bowls that could contain 4 sputa cups). At the end of the training, each participant was given a container for transporting sputum, 4 sputa cups, a presumptive case referral booklet and information, education and communication (IEC) materials to be used in educating clients.

Participants were instructed to conduct community education campaigns, identify presumptive TB cases in communities, collect sputum, and transport the sputum to the PHCs from where they were transported to designated laboratories. The results were sent back through the PHC TB focal persons, and all individual positive results were offered treatment at home or the local health post.

2. Cash incentives: The intervention Arm A (85) received cash incentives of 200 naira (USD0.78) for every presumptive case referred to the PHC for screening.

In addition, CHWs in the intervention arms received periodic supervisory visits by the research team, where their activities were assessed and encouraged and supported (financially) to carry outreaches.

The primary outcome of this study was initially the number of presumptive cases referred to the PHCs by the CHWs. However, TB case notification was used in the analysis instead of presumptive cases as many of the PHCs were new DOTS facilities and had no presumptive cases pre-intervention for comparison. The secondary outcome included the number of outreaches conducted per cluster. The outcomes were analyzed by cluster.

This study was initially designed as a panel survey; however, this was not possible as there was a large in-and-out migration of the PMVs, especially those who originally sent their apprentices to the training. At the endline, the same method used in recruitment at baseline was used, where all CHWs in the PHC catchment areas were targeted. This design was used to minimize sampling error, take into account the design effect, and prevent contamination across the three study arms. Hence, two independent cross-sectional surveys were used to obtain data from the CHWs.

Data on the number of outreaches carried out was collected from April 2019 to March 2020 during supportive supervisory visits. Data on the number of outreaches conducted between April 2018 and March 2019 was also obtained. Case notification data was accessed from the State TB program database. This included the number of TB cases notified from the selected PHCs between April 2018 and March 2019 (as pre-intervention Data) and between April 2019 and March 2020 (as post-intervention data).

Additionally, the study assessed changes in CHWs' knowledge of TB. A pre-intervention assessment was conducted using a pre-tested, self-administered structured questionnaire, standardized and validated by the national TB program for the 2017 TB Prevalence Survey in Nigeria. The questionnaires were administered just before the training workshop for study arms A and B and at the six PHC facilities for the control arm. A post-intervention assessment using the same tool was conducted after 12 months at the PHC facilities in each cluster. The results have been reported in another article.

All materials used for training and data collection have been made available in the Dryad and Zenodo repositories ²² .

Data was collected and entered into Microsoft excel spreadsheet, version 2013, then collated and analyzed using STATA version 13 (Stata, RRID: SCR_012763) and GraphPad Prism version 8 (GraphPad Prism, RRID: SCR_002798). The statistician was blinded to the study allocation until the data set was ready for final analysis. Descriptive data were analysed using summary statistics and presented in tables and figures. Effect sizes for outcome variables were calculated using odds ratios and 95% confidence intervals, and associations were tested using Chi-square (χ ² ) tests.

Ethical Approval was sought and obtained from the University of Uyo Institutional Health Ethics Research Board (UUTH/AD/S/93/VOLXXI/253). Approval was also obtained from the State Ministry of Health Ethics Review Board (MH/PRS/99/VOL.5/511). Individual verbal and written consent were sought and obtained from all participants. Verbal permission was also sought and granted by the Local Government PMV chairmen.

Only CHWs who consented to participate were recruited into the study. All presumptive cases identified were referred for screening and positive cases were linked up to the State TB program for treatment with DOTS. Following the study, CHWs in the control arm received the training.