How analysis of information can help health campaigns

Mr Chris Lukolyo speaks calmly, his deep voice holding the room in rapt attention. He is here, at the iHub, to demonstrate what open data can do and he manages adroitly with compelling, real-time visualisations. 

Pulse Lab Kampala, where he is the research and partnerships coordinator, is one of only three in the world, with the other two located in New York and Jakarta, Indonesia. 

They are part of the UN Global Pulse, an initiative of the UN Secretary-General that uses Big Data, or the analysis of massive datasets, to support the organisation’s humanitarian and development work.

Pulse Lab Kampala specialises in real-time data analytics. “What we are looking for is data that is being generated anyway (or unstructured data),” Mr Lukolyo explains. The lab does not collect data but works with partners who make data available to it. “I dedicate all my time not just to building relationships but nurturing them,” he says.

The results are as consequential as they are visually compelling.  In January 2013, there was an outbreak of measles in Uganda. Using mobile phone data from a 10-day period provided by Orange Uganda, Pulse Lab Kampala researchers were able to map the real-time movement of people into and out of Kampala city and Gulu Town in the north, and compare them to data on the spread of measles during the same period.

Mobile phone data is useful because the spread of infectious diseases depends on the movement of people. These methods provide a data-based basis for resource allocation during responses and can be used for other diseases once perfected.

Pulse Lab Kampala is also revealing what people in vulnerable groups are saying. By analysing talk shows and phone-in conversations on Ugandan community radio, researchers were able to determine public perceptions.

Using speech recognition technology, researchers could make both statistical and anecdotal analyses of certain important words or phrases in radio programmes, such as “outbreak” or “domestic violence”. The analysis revealed both the intensity with which ideas were expressed and the individual variations in meaning.

QUESTION OF CHANCE

Pulse Lab Kampala has got the government of Uganda enthusiastically involved in the project. So how did the government get on board?  “Sometimes it’s a question of chance,” Mr Lukolyo says. In Uganda’s case,  it happened that the then Health Minister Ruhakana Rugunda, now Uganda’s Prime Minister, was interested in the use of mobile technology to improve health.

Mr Rugunda learned about UN Global Pulse while serving as Uganda’s Permanent Representative to the United Nations and expressed an interest in his country hosting a laboratory. Pulse Lab Kampala began full operations in December 2013.

Initiatives like these need a champion, says Mr Lukolyo. “There is a desire, even among politicians, to do good for people and stay in power. We can appeal to that.

“If we can show you where the next outbreak is and you can deliver, your people will be very pleased,” he says, noting that not only governments stand to benefit from the control of infectious diseases. “The private sector has an interest in making sure the community in which they do business is healthy.”

Mr Lukolyo has some advice for professionals who may wish to work with real time data or who may have had trouble getting governments to hand data over to them.

“There has to be value to someone providing real-time access to their data. The partners need to see value for their data.” He suggests using historical data to prove that a concept works before making a case for real-time data.

“We make as much as we can publicly available and use as much open-source technology as we can,” he says. Pulse Lab Kampala takes privacy seriously, however. Data must be anonymised before use, which means all names must be removed. “As you leverage data, you have a responsibility”.

Mr Lukolyo notes that training and follow-up, particularly in government, is needed for long-term success, but argues that success goes some way in convincing sceptics.

“When we show visualisations to people, it’s easier to make a case instead of threatening ‘your health centre must deliver or we will imprison you’.”