Applying the iDARE Methodology in Uganda, Kenya, and Tanzania to Improve Health Outcomes During the COVID-19 Pandemic

Step 1: Identify Gaps in Health Outcomes

iDARE implementation in Uganda began in August 2020 when WI-HER, through the USAID Social and Behavior Change Activity (2020–2025), met with the Tororo District Local Government (DLG) and the regional implementing partner to review district-level data (DHIS2) to identify an area to begin work in improving health outcomes through behavior change interventions. DHIS2 showed that HIV-related health outcomes were the furthest from achieving their targets, with men (aged 20 years and older) and children (aged 19 years and younger) having the largest gap, specifically with the third HIV indicator, viral load suppression (VLS). The DLG reviewed the data and identified a facility to begin work. Table 1 shows Nagongera Health Center IV's (HCIV) clients who were virally suppressed (of those enrolled and active in care) at the time of the baseline. In September 2020, the iDARE coach supported the DLG to repeat this process and identify a second facility, Mulanda HCIV (Table 2), to begin work and support learning within the district among the 2 catchment areas. In contrast to Nagongera HCIV, Mulanda HCIV had a higher percentage of virally suppressed men and children active in care, while women had a lower percentage (84% suppressed) compared with 92% of Nagongera's actively enrolled adult female clients. Moreover, facility-level data revealed that in addition to VLS being an area that needed improvement for all clients, the total numbers of men and children active in care were low. It is critical to note that the iDARE teams were focused on behavior change of individuals in care and not service delivery improvement. Services continued to be regularly provided at the facility level throughout the iDARE implementation.

With sustainability and institutionalization planned at the project onset, initial iDARE teams were formed from existing personnel with key individuals identified and recommended by the appropriate local government structures, the DLG and the health facility's in-charge. Then, the teams recruited additional members from the community. Community influencers were selected through informal interviews with the target populations or a cohort (small group of nonsuppressed clients) to ensure the team(s) and overall process were community driven. The initial Nagongera HCIV iDARE team established a consenting cohort of 14 nonsuppressed men and 16 nonsuppressed children from the community who were actively enrolled in care. Mulanda HCIV followed suit, establishing an initial cohort of 14 men and 12 children. Both teams then met with the men, children (when they were over assenting age), and caretakers to understand gender, youth, and social inclusion root causes that may be linked to their behavior. For children, lack of adult support was identified as a barrier to antiretroviral therapy continuity and therefore, low levels of VLS. For men, work and timing for clinic visits led to low levels of antiretroviral therapy continuity and therefore low levels of VLS. In these informal interviews, the iDARE team sought to understand who influences cohort members' beliefs, attitudes, perceptions, opinions, and decisions that lead to a behavior. After identifying influencers mentioned by the various cohort members, the iDARE team recruited the influencers to join the team.

Step 2: Design Solutions to Increase VLS

The iDARE team coach supported Nagongera HCIV and Mulanda HCIV iDARE teams to analyze the findings from interviews with the cohort members to identify root causes of the behavior of the men, children, and the children's caretakers; document all root causes and prioritize issues to address; and brainstorm and design solutions to test and assigned individuals from the team to lead the implementation of the solution. Team leaders followed up with the responsible team members to check progress on a regular, agreed-upon basis.

Step 3: Apply Solutions and Assess Progress

iDARE teams in both facilities met to assess their progress on the agreed-upon solutions, review facility data, continue coaching through the iDARE process, and hold learning sessions. The iDARE team coach, along with iDARE team supervisors in local government structures, provided formal coaching sessions at least once a month. Because of COVID-19 regulations and safety measures, coaching was conducted virtually, in person where possible, and through a hybrid approach of virtual and in person.

The iDARE coach also conducted multiple in-person and virtual learning sessions for teams from Nagongera HCIV and Mulanda HCIV that allowed the facility teams to reflect and learn from each other on what worked and what did not and to stimulate new ideas that addressed common issues.

Step 4: Record Results and Document Lessons Learned

Table 3 shows successful and nonsuccessful solutions applied during Nagongera HCIV's and Mulanda HCIV's iDARE implementations. Learnings were harvested and shared between the 2 facilities at a learning session, which led Mulanda HCIV to test a Nagongera HCIV successful solution to address the lack of food for children for non-suppressed child clients at Mulanda HCIV. The iDARE team from Mulanda HCIV determined the solution was not a success because the religious leader was not as active in the community as the Nagongera pastor and was unable to mobilize food as easily. The Mulanda HCIV team further brainstormed to understand why the tested solution did not succeed as it did in Nagongera. First, the recruitment of the religious leader to the Mulanda HCIV team was done differently from that conducted by the Nagongera HCIV's team. In Nagongera, the pastor was identified as an influencer during the informal interviews with the 16 nonsuppressed children and their caretakers and asked to join the iDARE team from the beginning of implementation. In contrast, Mulanda HCIV specifically sought to find a religious leader as an influencer. They spoke to 3 caretakers of nonsuppressed actively enrolled child clients separately and asked who the influential religious leader was in the community; all named the same person. Health workers at the facility also identified the same individual when asked the same question. No children (active in care) or additional caretakers, beyond the previously mentioned 3, were consulted. The religious leader identified was asked to join the iDARE team and implement the successful solution from Nagongera; however, the religious leader from Mulanda was not as motivated nor as active as the pastor from Nagongera. This outcome serves as an example of how a solution must be appropriately adapted to the context and not duplicated. A potential way to adapt the solution in Mulanda could have been to ask 1 or more community influencers—identified by nonsuppressed children and where applicable, their caretakers—to lead the solution of mobilizing food within the community at a set time during the week.

Step 5: Expand Efforts to More Clients

The 2 facility iDARE teams in Tororo District expanded their efforts in multiple ways. In Nagongera HCIV: (1) after 2 months, the team expanded beyond their initial cohort of 30 nonsuppressed men and children active in care to further identify, design, and apply solutions for all nonsuppressed men and children (Table 3) who were actively enrolled in care; and (2) after 3 months, began to work on lost to follow-up among men.

In Mulanda HCIV, the team: (1) after 3 months, expanded their efforts to work on lost to follow-up among children; (2) after 5 and 9 months, respectively, expanded to improve VLS among active men and children by establishing 2 additional cohorts; and (3) after 6 months, with support from the DLG, identified TB case notification as another priority for iDARE application.

Results: Increased VLS Among People in Care

In 12 months, the iDARE team at Nagongera HCIV increased VLS among actively enrolled men in care from 65% (357/550) to 95% (534/562) and increased VLS among actively enrolled children in care from 60% (62/103) to 96% (100/104). Furthermore, they closed the gender gap among male and female children from 47% (22/47) and 71% (40/56), respectively, to 94% (48/51) of male children suppressed and 98% (52/53) of female children suppressed. In 11 months, Mulanda HCIV's iDARE team increased VLS from 85% (296/347) to 93% (324/350) among actively enrolled men in care and from 73% (38/52) to 96% (75/78) among actively enrolled children in care. A gender gap did not exist among children at Mulanda HCIV at baseline, and the iDARE team ensured that a gap did not emerge during the 11-month period, achieving 97% (36/37) male and 95% (39/41) female children suppressed from 74% (17/23) and 72% (21/29), respectively (Figure 2).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2

Percentage of Enrolled Men and Children Active in Care Who Were Virally Suppressed During iDARE Implementation, Nagongera Health Center and Mulanda Health Center, Tororo District, Uganda

Step 1: Identify Gaps in GBV Identification and Management

In May 2020, WI-HER, through the USAID Afya Nyota ya Bonde project (2018–2021), used program data to identify facilities with the lowest identification of, management of, and response for GBV survivors. All facilities supported by the Afya Nyota ya Bonde project were given GBV identification and management targets estimated by USAID and the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) based on the catchment area and the number of clients and facility size. GBV refers to harmful acts directed against an individual based on their gender.¹³ Although it disproportionately affects women and girls, GBV can involve individuals of both sexes and all genders. In Kenya, survivors are classified by sex, male and female. A GBV survivor is any male or female who has been or is being subjected to 1 or more of the 4 forms of GBV, namely, emotional, physical, sexual, and financial.¹³ The WI-HER team identified 8 facilities in 3 counties, Nakuru, Laikipia, and Kajiado, in Kenya with the estimated highest burden of GBV survivors, which accounted for 59.7% of the entire GBV PEPFAR target for all 419 project-supported facilities. After analysis of program data, the iDARE coach worked with 3 (of 5 project-supported) county government teams to identify up to 4 health workers per facility for training on all forms of GBV, World Health Organization (WHO)'s LIVES,¹⁴ and iDARE methodology. Before the training, these health workers submitted their facility-level baseline performance data from January to May 2020.

At baseline, on average, the 8 facilities had achieved a 4% average of their estimated facility-level PEPFAR targets, which is an aggregated target for males and females of all ages, including children. Furthermore, facilities and health workers only had the capacity and tools for the identification of, management of, and response for survivors of sexual violence. Health workers were not equipped with the knowledge, skills, or tools to identify, provide care, and/or report for survivors of emotional, financial, or physical violence. Additionally, across all facilities, there was extremely low identification and management of male survivors of violence, with an average of only 8 male survivors identified per month (Table 4).

WI-HER adapted existing Ministry of Health tools, approaches, and standard operating procedures to meet international standards and trained health workers on their usage. Furthermore, the training had extensive practical sessions on stigma and bias in GBV care and response. Health workers immediately began testing the tools in their facilities and provided feedback in real time to the iDARE team coach to ensure tools were practical and contextually relevant. Additionally, to ensure and measure GBV quality of care within facilities, WI-HER developed a coaching guide and GBV quality assurance tool to assess gaps in the provision of the minimum package of care for survivors of GBV. The GBV quality assurance tool uses direct observation; inquiry with providers and facility managers; and review of clinical records, guidelines, protocols, and documents to assess facilities based on the following domains: availability and appropriateness of services, facility readiness and infrastructure, case findings of survivors of GBV, survivor-centered clinical care and provider-survivor communications, forensic examination and handling of evidence, referral system and follow-up with survivors, reporting and information systems, training and quality improvement, and health care policy and provision.

Initial iDARE teams in Kenya were formed with the county government GBV focal person and health facility in-charge identified and recommended by the appropriate local government structures and health workers at the facilities.

iDARE work in facilities focused on service delivery improvement. The initial iDARE team also included health workers who served the 3 critical functions of the continuum of care for GBV survivors: (1) identify and screen cases (generally stationed in the outpatient department); (2) provide GBV counseling and psychosocial support; and (3) clinicians (nurse/doctor/clinical officer) who conduct full physical and, when appropriate, gynecological examinations. In most cases, a health worker served more than 1 of the functions. The virtual training was conducted over 6 weeks with low-dose, high-frequency training and coaching sessions. The health workers were trained on all 4 forms of GBV, expanded GBV tools, simplified WHO LIVES,¹⁴ and iDARE. The trained health workers then identified additional team members within their facility to support their iDARE work. Most facilities recruited at least the health information officer as well as representatives from other departments to improve identification throughout the facility.

Step 2: Design Solutions to Strengthen Capacity in GBV Identification and Management

Initial iDARE teams designed early solutions to test, along with plans to recruit additional members to the iDARE team, during their 6-week virtual training. As the teams expanded, they designed additional solutions to address gaps they identified with colleagues from other departments in their facilities. Furthermore, the iDARE team coach engaged county government officials to support the institutionalization of the work to improve GBV identification and management in the 3 counties. This resulted in 1 county identifying and establishing a GBV focal person for a facility that did not have one.

Step 3: Apply and Assess Solutions Tested

The iDARE team met regularly to assess facility progress on the agreed-upon solutions, review facility-level data (e.g., the GBV register), continue coaching through the iDARE process, and hold learning sessions. WI-HER, along with iDARE team supervisors in local government structures, provided formal coaching sessions at least once a month. Coaching was conducted virtually, in person where possible, and through a hybrid approach of virtual and in person due to COVID-19 regulations and safety measures that were in place.

Strengthening capacity ranged from reorienting teams and new team members on the expanded GBV tools, including the GBV checklist and quality assurance tool, or conducting training on the forms of GBV and how to identify, manage, and respond to a survivor based on the minimum package of care. The iDARE team coach supported the teams to identify and record additional gaps and barriers they experienced, brainstorm new solutions to begin testing, and supported the teams in the quantitative and qualitative data recording and analysis.

Due to the distance between facilities and COVID-19 movement restrictions during lockdown, learning sessions took place over Zoom. While a learning session has the greatest impact when all team members can participate, in Kenya the facilities were heavily affected by COVID-19 and could not afford to have multiple team members away from their stations even for 1- or 2-hour sessions. Therefore, Kenya teams suggested having only their iDARE team members participate in learning sessions and report back to their wider team. Although recordings of the learning sessions were shared with all team members, the use of Zoom was less interactive and engaging than in-person learning sessions. Data and network challenges caused issues in sound and overall participation. It was also difficult for the health workers who could not attend to find time to download and review the recordings to hear the learnings from other facilities.

Step 4: Record Successful Solutions in Improving GBV Identification and Management

Table 5 shows identified gaps in GBV care and the tested successful sample solutions recorded in a provincial general hospital–county referral in Kenya that was not an early adopter of the iDARE methodology or improving GBV identification and management. However, once the hospital began to see early results showing improvement, it became more confident in implementing iDARE and greatly improved GBV identification and management of female and especially male survivors.

Step 5: Expand Awareness of Availability of GBV Care

iDARE teams worked on expanding their efforts in the community to ensure that individuals who have been subjected to violence know there are support services available to them at the facility (Table 5).

Results: Increased GBV Identification and Management

In 10 months of implementation, the 8 facility-based iDARE teams increased GBV identification and response by 641.29% for male and female survivors of all 4 forms of violence. From June 2020 to March 2021, a total of 5,517 cases of GBV were identified, with an average of 551 persons per month. The 8 facilities increased identification and management of female survivors from an average of 81 per month (baseline) to 364 female survivors per month (June 2020 through March 2021). Additionally, the facilities increased identification of male survivors from an average of 8 survivors per month (baseline) to 188 male survivors per month (June 2020 through March 2021), resulting in a 2,250% increase in male survivors per month identified and managed. These results demonstrate the progress the facilities made, despite the COVID-19 pandemic and a health worker strike in December 2020 and January 2021 (Figure 3).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3

Gender-Based Violence Cases Identified and Managed by iDARE Teams, by Sex, in 8 Facilities in Nakuru, Laikipia, and Kajiado Counties, Kenya, January 2020–March 2021

Step 1: Identify Gaps in Coverage of MDA

In January 2021, WI-HER, through the USAID Act to End Neglected Tropical Diseases—East Program (2018–2023), began supporting the Tanzania National NTD Control Program (TZNTDCP) to review national NTD program data, including surveys evaluating coverage of MDA for onchocerciasis and trachoma treatment to identify a district to begin testing iDARE to improve coverage for frequently missed or refusing populations. The national GESI assessment identified men who take alcohol and school-age children as the most commonly missed or refusing groups for community and school MDA, respectively.

In March 2021, in the Pangani district, the TZNTDCP worked with district council leadership to use existing data (Table 6) and conduct community-based informational interviews to identify GESI challenges for community and school MDA access and uptake of the deworming medication albendazole. To explore low access and uptake among both types of MDA, the implementation team sought to understand if those targeted during community MDA, specifically adults, influence their children's behavior during school MDA. Data were collected from the MDA registers by the district NTD teams during the pilot. From the above information, 13 schools in 2 zones—a group of wards, which constitute a group of villages—showed an average gap of 20% for both female and male school children in the available disaggregated data. Two schools in each of the Mwera and Mkwaja zones that had the largest absolute number of missed school-age children were selected for application of iDARE to improve access and uptake of MDA.

Cohorts were formed in the 2 zones and represented various groups to understand the reasons for missing the last MDA (community, school, or both). For community MDA, reasons included people were not at home during MDA because of the nature of their livelihood activities (e.g., migration for farming, pastoralism, and fishing); households were too far for the community drug distributors to reach; fear of side effects, especially when mixing with other drugs for chronic diseases such as hypertension and diabetes; or the common belief among both men and women that MDA medicines could cause impotency and/or infertility. Reasons children missed or refused school MDA included school absenteeism; culture and traditions of the community; lack of food, which caused side effects and then fear; parents not allowing their children to take medication; fear of infertility; general fear; and discomfort in taking medication.

Step 2: Design Solutions to Address Root Cause Affecting Access and Uptake

Initial teams were formed with key individuals identified and recommended by the appropriate government structures, including the NTDCP focal person, regional NTD representative, and district stakeholders.

The iDARE work in Tanzania focused on SBC. The initial iDARE team interviewed children, caretakers, and teachers on the various general and gender, youth, and social inclusion root causes affecting behavior, as well as the influencers in their lives. The influencers were then asked to join the iDARE team to support the behavior change work. The iDARE teams continuously expanded action plans before and during implementation with various members supporting the iDARE work, including supervisors and iDARE team members. iDARE team members in Pangani Council documented the identified barriers, gaps, and issues affecting MDA access and uptake and their designed solutions as well as results. It was adapted to the local language to be used by all team members.

Step 3: Apply and Assess

Although the implementation period was short in Tanzania, the iDARE teams still met regularly to assess their progress on the agreed-upon solutions through the review data (document their solutions, measure intention among cohorts, and collect final uptake data from the MDA register), and receive coaching support from local government and iDARE team coach.

The iDARE team coach, along with team supervisors in local government structures, provided in-person coaching sessions throughout implementation. COVID-19 delays caused the length of the planned implementation to be shortened from at least 3 months to only 1 month; coaching was then provided intensively throughout the entire duration and in person. COVID-19 restrictions and time constraints prohibited additional coaching visits in Tanzania.

Step 4: Record Progress Toward Intent to Take Medication

The teams in the Pangani district determined a proxy indicator, intent to take medication, as a means to measure progress of the intended outcome: uptake during the upcoming school MDA. Because of limited implementation time before the school MDA, only 1 measure of intention was taken after the team began testing solutions (Table 7).

In addition to the above-mentioned groups, iDARE teams also assessed community members in Mkalamo (15) and Mwera (7) to gain further insight into the communities' outlooks on MDA. These individuals were selected based on if they did not take the medication during the prior MDA. Twelve of 15 community members in Mkalamo and 7 of 7 community members in Mwera stated they intended to take medication at the next MDA. Additionally, as nonenrolled school children previously could not be verified by school MDA registers regarding whether they had taken medication, the iDARE team intentionally identified a group of nonenrolled children in Mkalamo to understand their intent to take the next MDA. All 24 of the nonenrolled school-age children intended to take the medication.

Step 5: Expand to Additional Districts

After the successful improvement of school MDA coverage, the NTDCP held a reflection meeting on learnings from the Pangani District pilot and began plans to scale up to 6 additional districts to improve community and school MDA access and uptake for onchocerciasis and trachoma treatment.

Results: Improved School-based MDA Coverage

After 1 month of implementing iDARE, the Pangani district achieved 99% coverage in the 4 pilot schools (Table 8). Of those nontreated at Mwera Primary School, 4 registered children had transferred to another school, 3 children were sick during MDA, and 20 were absent from school. In Stahabu Primary School, 1 child who was registered transferred to another school before MDA, and 1 child was pregnant and not offered the medication. iDARE teams have reviewed these reasons for missing MDA and have begun to design new solutions to improve the next MDA when the time comes.