Smart data analysis saves lives in Mozambique

Bio-Statistics and Modeling PhD project results in recommendations with lower mortality rates

It seldom happens that the social relevance of a PhD project is so undeniably clear as in the case of the Bio-Statistics and Modeling project of three students from the Eduardo Mondlane University in Mozambique. By analyzing data on mortality causes such as HIV/AIDS, the team was able to draw conclusions that help the health sector save lives.

Osvaldo Loquiha leads the team of three, which also includes Adeline Martins and Adeline Juga. They are all active as lecturers and researchers at the Science Faculty, Department Mathematics and Informatics, at the Eduardo Mondlane University in Mozambique. They work in close collaboration with the Center for Statistics at UHasselt.

They can use these recommendations as a tool to make better decisions and actually reduce mortality rates.

Osvaldo Loquiha
Lecturer and researcher at the Science Faculty, Department Mathematics and Informatics, Eduardo Mondlane University, Mozambique

Analyzing and processing public health data

Osvaldo Loquiha explains what the PhD project is all about. “Simply put, the project is about analyzing and processing public health data, in order to draw conclusions and make recommendations to the health sector in Mozambique. Some of these recommendations will without doubt result in lives being saved, directly or indirectly. As an example, the team examined the effectiveness of government awareness campaigns on HIV/AIDS.”

Data sources

The team gets its public health data from several different sources. The Ministry of Public Health in Mozambique is the most important data source, but the team also used data from the ‘INSIDA 2009’ survey on HIV in Mozambique as well as additional data from collaborations with individuals that are involved in other research programs at the university.

Looking for hidden information

“We analyze the data using different methodologies, and search for statistical and mathematical relations and structures. We also try to combine data streams, in order to reveal tendencies and hidden information. And of course, we are always looking ways to optimize the extrapolation of our data and findings.”

“Our purpose is to draw the correct conclusions and present these as recommendations to the Ministry of Public Health as well as to other health care organizations in Mozambique. They can then use these recommendations as a tool to make better decisions and actually reduce mortality rates. As an example, we found that the mortality rate in Mozambique is highly dependent upon geographical location. To be more specific, there is a clear correlation between the mortality rate and the proximity of a city. The greater the distance one lives from a city, the higher the mortality rate. If a facility in a small town doesn’t know how to treat the patient, or if they don’t have the right equipment, they have to transfer the patient to another location. The mortality rate changes in direct proportion to the distance to the location with better facilities.”

Improved methodologies

The team developed improved methodologies to cope with public health data issues, and is actually using this experience today to help the Ministry of Public Health improve the quality of their data.

“In the future, our university will be able to use our methodologies in other projects as well. It will help the university to work its way through a great deal of data and analyze it faster and in greater depth.”

Three reasons to choose SAS

SAS was chosen as the main tool for this project for three reasons. “First of all, we already had experience with this software, during our studies at the UHasselt in Belgium. Hence, we did not have to start from scratch. Secondly, SAS is easy to use and easy to program. This enables us to concentrate on the core of our project, and lets the program do the work for us, instead of the other way around. Thirdly, the SAS solution uses very stable algorithms. It’s also very extensive and has many cross reference possibilities. I especially like the ‘NLMIXED’ procedure. This is a SAS procedure that fits nonlinear mixed models. In our research we mainly applied it to model overdispersed count data. Its ‘likelihood’ function proved to be extremely useful in this project.”

“Another big plus is the ease in writing SAS macros. We can use them for this project, but we can also transfer them to future projects and even hand them off to other individuals involved in other research.”

Teaching (with) SAS

Loquiha states that he will keep on using SAS in future projects. He used to work with free software in the past, but was deterred by the lack of stability. Today, he uses SAS as one of the main software tools for teaching. “In my lessons on ‘Multivariate Statistics’ and ‘Regression Analysis’, I teach my students how to use SAS.”

eduardo-mondlane-university-mozambique

Challenge

Analyzing data for a Bio-Statistics and Modeling PhD project at the Eduardo Mondlane University in Mozambique.

Solution

SAS® Analytics

International collaboration program

This PhD project is part of an international VLIR-UOS collaboration program with different Belgian universities. The University of Hasselt is leading the project Bio Statistics and Modeling, but universities from Ghent, Antwerp and Leuven are also involved. VLIR-UOS supports partnerships between universities and university colleges in Flanders (Belgium) and the South (Africa, Asia, Latin America) looking for innovative responses to global and local challenges. The institutional cooperation with Eduardo Mondlane University (UEM) via the IUC-programme, nicknamed DESAFIO, is one of these partnerships, well established and structured over the years.

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Back to Top