Equity Case Study: Zambia - Using mobile phones to improve early infant HIV diagnostic services, post-natal follow-up and care
By Abhijit Shanker
04 November 2011 - Long delays averaging 6.2 weeks for infant human immunodeficiency virus (HIV) test results pose a significant barrier to mothers and families seeking equitable access to early infant initiation of antiretroviral therapy, especially in rural settings in Zambia. Such delays contribute to loss of follow-up and possible death of 30 per cent of affected children if no interventions are provided. Project Mwana, a mobile health (mHealth) initiative implemented by the Zambian Ministry of Health (MoH) with support from UNICEF and collaborating partners, Boston University affiliate The Zambia Centre for Applied Heath Research and Development (ZCHARD) and the Clinton Health Access Initiative (CHAI), aims to reduce this delay and improve children’s chances for a healthy life.
Project Mwana focuses on finding appropriate, scalable and impactful ways through which mobile technologies can strengthen health services for mothers and infants in rural health clinics. The technology utilizes RapidSMS, a free and open-source framework for building mobile applications for dynamic data collection, logistics coordination and communication, leveraging basic short message service (SMS) mobile phone technology. Project Mwana went live on the 14 June 2010, and to date it has relayed more than 3,000 infant HIV test results with reduced turnaround times of around 50 per cent, having a great positive impact on rural facilities.
Since the project’s inception, UNICEF has provided the MoH with the strategic vision and support necessary to develop the software, outfit some of the hardware technology and make available additional human resources to strengthen capacity for software and hardware management. As the project scales up, UNICEF will continue to provide support in these areas, including capacity development at national and district levels. In 2010, Project Mwana was piloted in 31 clinics within six provinces across Zambia, and it is currently beginning a national expansion.
Large numbers of infants in Zambia are infected with HIV in utero, at delivery or when breastfeeding. Among the infants who contract the virus from their mothers, about 30 per cent die before the age of one and 50 per cent before the age of two, if no interventions are provided. These deaths contribute significantly to the national under-five mortality rate. Survival rates are up to 75 per cent higher for HIV-positive newborns that are diagnosed and begin treatment within their first 12 weeks of life. However, many children born to women with HIV are not being systematically monitored, and many are identified as infected with HIV only when they become very sick.
Exposed infants need a special type of HIV detection test, which differs from the standard antibody test for adults and older children because infants still have their mother’s antibodies in their bloodstream. Only three laboratories in Zambia possess the capability to perform the required test. Transport of the samples and the test results, especially for rural clinics, is therefore one of the biggest obstacles facing early infant diagnosis and initiation of Anti-retroviral Therapy (ART) for HIV-positive infants, significantly reducing their chance of survival. In June 2010, the MoH, in partnership with UNICEF, CHAI and ZCHARD, launched the Project Mwana RapidSMS pilot to reduce delays in transmitting results from the HIV test laboratories to the rural health facilities via SMS message. To date, the project has been piloted in 13 predominantly rural districts of Zambia.
Strategy and Implementation
UNICEF, ZCHARD and CHAI collaboratively initiated Project Mwana, developing slightly different tools and approaches - all focused on shortening the turnaround times for HIV test results - and launching three separate, but coordinated pilots, to explore the method’s efficacy. The mobile solutions developed for Project Mwana were created with specific health objectives, aligned with the national health strategies for Zambia. The primary project objective of strengthening early infant diagnosis aims both to increase the number of mothers receiving results and to reach mothers in a faster, more efficient manner using the “Results160” SMS application. A second objective is to improve the rate of postnatal follow-up, increasing the number of birth registrations for clinic and community births, while also raising the number of clinic visits for mothers through community-health worker tracing using the RemindMi application.
The system relies on communication among rural health workers in remote clinics and community-based agents supporting health services in rural communities. These participants own mobile handsets, which they use to send and receive Results160 and RemindMi messages via SMS.
This system design comprises three fundamental components:
1. End users who enter and receive data through their phones (clinic and community-based health workers using SMS), their computers (lab workers over the telecom data network), or their printers (clinic workers in busy labs with SMS printers).
2. A centralized server sitting in the Ministry of Health connected directly to the telecom providers, which securely coordinates all the data that flows through the system and directs it to the correct recipient.
3. A password protected web interface for Ministry of Health officials and implementing partners, where they can view information, graphs, and maps showing how the system is performing in real-time.
The facilities are located in rural, peri-urban and urban centers, with the majority of them in the hardest to reach rural areas. The tools were developed primarily to address the needs of these remote clinics. The solutions were developed by local and international software developers in an iterative fashion, and they were tested and modified in consultation with health and community workers. The tools were also developed with the intention to scale up their use after the pilot program, in consideration of issues such as maintaining privacy and keeping costs sustainable. They were developed to provide real-time monitoring of potential bottlenecks in the transfer of the samples, to create systems of accountability among the provincial and district staff, and most important, to help with the patient-tracing tasks of persuading mothers to bring their children to the health center for the dry blood spot (DBS) test and to return for the results.
Progress and Results
Evaluation. The preliminary monitoring and evaluation analysis of Project Mwana sites indicates that:
>>SMS delivery of results can increase turnaround times by 50 per cent on average, with a greater positive impact in rural facilities.
>>The birth registration component of the RemindMi application has been adopted enthusiastically by community health workers, registering the majority of births in their catchment areas.
>>It is too early to know if SMS reminders have significant impact on increasing adherence to post-natal appointments.
>>It has not yet been established whether the faster turnaround time for results leads to earlier ART access for exposed infants.
>>ZCHARD is in the process of conducting a formal evaluation of the system, looking at its quantitative impact, at the qualitative experience of stakeholders, and at outcome level indicators.
Expanding Project Mwana’s Reach. Zambia is in a position now to start scaling up the system. A national scale-up plan has been developed, commencing with a preparation phase and then shifting to an iterative phase where clinics are trained and added to the system and the problems and successes of the additions are evaluated. The aim is to achieve national scale by 2013, with health facilities offering early infant diagnosis services. The preparation phase will focus on solidifying the technical, physical, monitoring and human infrastructure to allow the system to handle the stresses of scale. Throughout the scale-up process, the project will be closely monitored to ensure the systems are having a positive effect on the targeted health challenges.
As with any large-scale project, there are a variety of risks that can impact the success of scale-up, usually identified as one of three types:
>>Project risks are associated with the technical and implementation aspects of the scale-up;
>>Process risks are associated with the processes, procedures, resources, communication and operation of the scale up; and
>>Scale-up risks are associated with specific risks tied to the expansion of the system and process of scaling.
The project team confronted two main challenges during the building of the RapidSMS systemsboth of them categorized as process risks: ownership of the project prior to initiation, and coordination among the partners. Approaches used to overcome these challenges were to meet and strategize regularly during the pilot phase and to ensure that there was only one coordinating body (an SMS technical working group led by the MoH), one joint plan, and one common monitoring and evaluation framework. Although the three pilots were slightly different, a common goal was cultivated and shared with the MoH showing that partners were moving together toward a shared vision.
The following principles related to scalability and sustainability have been identified to guide the scale-up of Project Mwana:
1. Government leadership
>> Involve the MoH from the beginning
>> Integrate the project into long-term planning
>> Leverage the national health eco-system
>> Integrate data into district reporting
2. Locally sourcing
>> Employ a permanent local software development team
>> Have a permanent project manager who can coordinate partners
>> Create government-led working groups
3. Cost control
>> Negotiate with telecom companies for scale, not pilots
>> Utilize the phones people have rather than purchasing and supporting a national phone system
>> Create district-level training teams
>> Make decisions based on identified needs of the end users
>> Create the tools with the people who are going to use them
>> Test early and often; don’t worry about failing and stay adaptable
>> Use open source tools that can be customized to local needs
SMS technology is a powerful innovation that in Zambia has reduced delays in receiving early infant diagnosis (EID) DBS HIV test results, improved communication among health care providers and community volunteers, and more important, encouraged patients to return to the clinic for their test results with greater confidence.
With national mobile phone penetration growing at an unprecedented rate in Zambia, there is an enormous opportunity to utilize this platform to scale up support for recognized health interventions and to explore other areas of need. Once the preparation phase is completed, scale-up will require a phased approach to systematically add more sets of clinics and monitor how they are using the system, keeping two main objectives in mind:
- To improve the turnaround times of test results delivery by 30 per cent to 60 per cent, and to increase the number of HIV-positive infants receiving ART treatment within the first year of life; and
- To improve oversight of the test system and to strengthen linkages among health facilities, community health workers and caregivers.
The experiences from development of the Project Mwana system can serve as the basis for future mHealth projects. The system could potentially be extended to maternal and child health areas such as the prevention of mother to child transmission (PMTCT) of HIV and nutrition, as well to other results delivery and diagnostic mechanisms or national health programmes for women and children. 6
Mobile phones, which can be found even in the most remote areas of Zambia, show the potential to “reach the unreached,” i.e., the most disadvantages communities. As the Project Mwamba RapidSMS pilot reveals, these powerful tools can effectively empower the public and remote health workers with diagnostic and information-sharing capabilities, provide training, and ultimately improve monitoring of activities and outcomes.