Skip to main content

How Power BI enables remote communities in Africa to track AIDS

The Elizabeth Glaser Pediatric AIDS Foundation (EGPAF) seeks to end global pediatric HIV/AIDS through prevention and treatment programs, research, and advocacy in 19 countries worldwide.  The number of new HIV infections among children has declined more than 90% in the U.S. and 70% worldwide. Over its 30-year existence, EGPAF has played a major role in these dramatic reductions in India and many African countries. EGPAF partners with Ministries of Health (MOH) to provide targeted technical assistance to clinicians and to generally strengthen health systems.

EGPAF is a data-dependent organization that relies on timely and accurate data for decision-making. Improved access to and use of data will ultimately save lives, and to accomplish this, the organization is in the process of deploying Power BI globally. For example, this great Power BI dashboard shows the impact EGPAF is having across the globe.

EGPAF also uses Power BI to communicate their achievements to their sponsors using data stories. This Very Early Infant Diagnosis (VEID) data story is a great example.

One of the goals for the EGPAF organization is to enable every employee of EGPAF around the world to make use of the data collected in every country and make informed decisions on this without having to wait for it to be collected centrally. Data collected in Tanzania is different from data collected in Kenya, and EGPAF employees in each country have slightly different questions. To centralize all the data and answer all the different questions is hard. To enable the local teams to make these decisions EGPAF has rolled out Power BI in every country. EGPAF central BI team members Stephanie Bruno and Shannon Lindsay, located in the US, also run Power BI user groups in Washington DC and Pittsburg. In this Power BI community blog post, Stephanie and Shannon describe how Power BI can literally save lives.

As part of their involvement with the PUG they came across a program run by Microsoft called MySkills4Afrika, which aligns well with their own goals. The MySkills4Afrika program allows “Microsoft employees from all over the world volunteer their time, talent and considerable expertise to support the company’s mission to empower every individual on the planet.” EGPAF signed up for this program to organize Power BI training in Africa for their local data analysts, often experienced Excel users. The MySkills organization agreed to host 3 trainings in Africa delivered by members of the Power BI team: Patrick LeBlanc in Lesotho, Kasper de Jonge in Kenya and Maggie Sparkman in Cote d’Ivoire.

Lesotho

The Lesotho training took place in Maseru, Lesotho.  17 employees attended: 14 from Lesotho, 2 from Malawi and one from Mozambique.  We spent five days digging into Power BI, but the best part of each day were the excursions that we would take assisting in solving real EGPAF problems.

During the week, we encountered many challenges.  For example, the instructor’s adapter to connect to the projector was broken.  You would assume that you could just run out and purchase one at the local technology store.  Unfortunately, not in Lesotho.  Instead, a loaner laptop was provided, and all instructor content had to be moved to that laptop.  We encountered many other technical challenges, but fortunately the IT team attending the training quickly solved them all.  We encountered many other challenges around data quality but using Power BI and a little data cleanup we easily overcame those challenges.  In the end, we were able to solve many problems while providing a solid foundation of assisting in improving EGPAFs data culture.

Nairobi, Kenya

The Kenya training took place in Nairobi, Kenya for 30 EGPAF employees from Kenya, Tanzania and Uganda. We took them through the entire Power BI platform in a week, from visualizations to DAX and modelling. Most of the content for the training was based upon the freely available content you can find here.

It was amazing to hear the stories of how hard even the simplest tasks can be. For example, collecting data happens at sites in the very remote places  fully by hand, often no power is available, nor wireless internet or a mobile data plan. This data is collected by “data collectors” who go around to all these sites to record the transactions into an excel spreadsheet with their laptops. There are a few obvious problems with this method. First, the data quality is not guaranteed. Also, each country has its own way of collecting data. We spent considerable time on Power BI data flows , as each team might be able to load their data into a central data lake for everyone to use. The EGPAF team expects this technique to help significantly on their journey to better data quality and improved data reuse.

At the end of the training there, we also had two days to go on a safari through the Amboseli natural park where we saw hundreds of elephants, cheetahs, impala, buffaloes, lions, and much more.

All in all, an amazing experience. It’s gratifying to see Power BI being used for scenarios to change lives at a fundamental and concrete way.

Cote d’Ivoire

The Cote d’Ivoire training took place in Abidjan, Cote d’Ivoire for 24 EGPAF employees from Cote d’Ivoire and one from Congo. The attendees were at varying levels of proficiency with Power BI, from those who had never seen it to those who had used Power BI Desktop a few times to create reports.

Perhaps the part they liked best was that the training and all the training materials were in French. All of them have some proficiency in English but feel more comfortable reading and speaking in French. In the pre-training survey, 100% said they preferred communicating in French in the workplace. 91% (20 out of 22) have a French operating system on their computers and phones, and 95% (21/22) prefer reading documentation and slides in French. Thus, the French classroom environment gave them more confidence to speak up and participate.

The first day went like a typical class – slides, presenter demo, then everyone doing the exercises at their own pace. At the start of the second day, the attendees had this feedback: They had a hard time following the demos, and they didn’t like going at their own pace. They wanted to go through the demos along with the presenter, and they wanted work through the labs together, step by step. Even the more advanced students wanted that. So we did. They did the demos with me, and then we all did the labs together, step by step. The more advanced students helped their neighbors. The pace was slower as a result, but the students were more engaged – they had to keep up or make the whole class wait. In the end, they understood the material better. Even so, in the survey after the training five of the 15 respondents still thought the pace was too fast.

The last day, we started with an extract of data in the format of one of their standard data sources, DHIS-2: a PivotTable in Excel. We unpivoted the data and added a column to group the ages by child, adolescent, and adult. They were excited to see how to get their data into a usable form.

The attendees filled out a survey before and after the training, and the results paint a picture of a successful training. For basic Power BI tasks, the number of votes of feeling ‘not at all comfortable’ dropped from 67% before to 4% after, and the votes of ‘very comfortable’ rose from 7.4% to 40%!