Insights from using Interactive Voice Response to train Community Health Volunteers on COVID-19 in Kenya

By Christine Odour, Program Manager, Digital Health and Monitoring at PSI

Remote training and capacity building for healthcare workers has been a critical response mechanism to the COVID-19 pandemic. In a PSI article on “Powering Remote Capacity Building of Healthcare Workers using WhatsApp” published in December 2020, we shared insights on remote learning for private sector health providers focused on COVID-19. In this article, we share our key learnings on implementing remote learning on COVID-19 with a different cadre of the health workforce, Community Health Volunteers (CHVs), in Kenya.


Community-based health workers have a vital role to play in preventing, detecting, and responding to COVID-19. They can: ensure sensitization to create awareness at the community level; dispel myths and misconceptions about the disease; support screening efforts, and refer suspicious cases – particularly in rural, remote, and vulnerable communities. These critical healthcare workers need access to high-quality and easy-to-use information that will help increase their ability to engage the general population at the community level, about the virus.

Having successfully implemented remote training for health providers using WhatsApp, PSI and PS Kenya in collaboration with Viamo developed remote training for CHVs using the Interactive Voice Response (IVR), a technology that enables the target audience to interactively engage with automated information delivered through the use of voice. IVR can be used by any mobile handset, and is relatively easy to use. The user either dials in or gets an automated call from the IVR line, and the information is provided using voice recordings, with easy prompts for the user to navigate the information. The majority of the CHVs have basic phones and limited access to internet, therefore the training content for CHVs needed to be delivered in a simple fashion given their lower capacity. IVR provided a potentially effective channel to meet the CHV needs.


The IVR training package entailed 6 modules and a post test. These modules were recorded in English with each module running for approximately five minutes. The CHVs received notification via SMS on the learning schedule a day ahead of each lesson with the assumption that they would create time to answer the 5 minute call. The modules focused on various aspects of COVID-19 that would be relevant to the CHVs and their target audience in the community.


Approximately 1,000 CHVs across 14 counties in Kenya were targeted for remote training, activated through an automated call. The call pick-up rate was above 85%, with approximately half of the participants completing the lesson, and 1 in 3 participants proceeding to the post test quiz. The graph below highlights learner engagement rates across all modules.

COVID-19 CHV Training Modules


Based on the post learning quiz, learner performance varied by module as presented in the illustration. CHVs performed better on modules focused on containment of the pandemic through behaviour change, as well as on identification and  monitoring of suspected COVID-19 cases (See modules 3 and 4). However, CHVs performed poorly on facts about COVID-19 (See module 2).

Graph illustrating quiz correct response rate by module


In order to unpack learnings on the uptake and performance of CHVs, we conducted a follow up qualitative survey with a sample of CHVs involved in the intervention to gain insights on what their experience was with the platform, including some of the possible drivers on below-par uptake and performance.

Duplication of efforts: Most of the respondents mentioned the course was moderately helpful to them since they had completed other COVID courses from other organizations they work with. They mentioned fatigue from high requests to participate in trainings during COVID-19 pandemic. There is need for better coordination between partners involved with supporting CHVs to ensure that adequate capacity building is provided, minimizing duplication of efforts.

Language challenges: The content was developed and delivered in simple English, given that the CHVs had received at least elementary level education taught in English. However most respondents indicated they did not speak English as their primary language therefore it was difficult to understand. Language considerations should be effectively taken into account as part of the design process, with focus on understanding what languages users are most comfortable interacting with.


Prioritization of training: Despite the bite-size approach to training, respondents complained of not having adequate time to learn due to competing priorities or did not schedule their learning time well—nor were they able to call back to continue learning. Efforts should be made to ensure the target audience are aware of the value of the training so as to prioritize their time. Mechanisms to incentivize participation should also be considered, be it financial or non-financial, given that it is a key driver of participation for in-class training, and often tends to be overlooked with remote training.


Phone related issues: Respondents highlighted having issues with their phone battery dying midway through the course. Further to this, they experienced competing priorities for phone, including receiving calls mid-way through the session. Effort should be made to provide guidance to participants regarding embarking on the training when they are likely to have the fewest interruptions with phone calls, as well as ensuring they have sufficient battery life on their phone.

"It was quite difficult understanding English used during the training; we mainly speak in Kiswahili even with people we reach with information during our work. It would be helpful to conduct these trainings in Kiswahili to enable us also to know the right terms to use when engaging with the public during our work."

- Respondent

"I received a notification on the lessons ahead of time, however some urgent issues came up during the scheduled time and on the day of learning which made me not complete the lesson."

- Respondent

"The nature of our work entails receiving a lot of phone calls from the community members or from the organizations we work with alongside our personal calls. This created interruptions during the lesson…My phone battery also ran out."

- Respondent


Understanding the dynamics around the enabling environment for learning through the IVR platform for CHVs is key, as Kenya moves into the second and third wave of COVID-19, which presents new learning requirements for CHVs. We have learned that engaging the learners through a language they are most comfortable with is of high priority. All learning scripts will therefore be translated to Kiswahili, the national language alongside other community languages where needed.

In addition, using voice exclusively during learning may be monotonous especially for learners who have already covered some of the content at hand. We intend to explore gamification as an alternative, more interactive mode of learning to sustain learner engagement.

These insights are helpful in our iterative approach to improving our digital health solutions with end users at the centre of our design process.

Sign up to
Receive Updates

Donate to
Support Our Work