Skip to 0 minutes and 13 seconds CHRIS GRUNDY: During the outbreak, we’re collecting huge amounts of information. And that information, a lot of it has some form of link to location. So with our case information, we’re locating it through village, that case comes from and the district that the village is within. At the same time, we’re collecting lots of information around the geography. So we’re mapping whereabouts are village locations are. We’re mapping where the district boundaries are. We’re mapping where our clinics are, where our services are. And using those, we can bring those two pieces of information together to allow us to the produce maps, to produce tables which show how the outbreak is developing and what is happening where.
Skip to 0 minutes and 55 seconds One of the biggest jobs is cleaning up the case information. So when a case comes into a clinic, it’s recorded in what we call a line list. And that will record the information about where that person lives, where they were when the symptoms started to show, and any of their contact information. That’s a massive data set and requires lots of cleaning up. And one of the parts of that cleaning up is ensuring that the village that they specify or the locations that they give, that we can actually put those on the map. So one of the big jobs is to look at whether or not the villages that are listed appear in our data set.
Skip to 1 minute and 32 seconds Especially in this particular outbreak, contact tracing is particularly important. So lots of work goes into looking at where our contacts are, can we track those contacts down, can we follow up those contacts? The location that this outbreak is taking place, there was no map data originally. And all of that had to be generated at the start of a process. So we have to map the administrative areas, locate the villages, get a road network for us to use. Normally, this would have taken a massive amount of time. It would have taken years if we’d done it using very traditional methods. But with current availability of satellite imagery, we can bring in crowd-sourcing to map an area very, very quickly.
Skip to 2 minutes and 15 seconds So instead of one person walking around or driving around a road network, we have thousands of people all over the world sitting in front of their computers tracing features into their computers, which gets put onto a central database. Computer software is getting better to work in the field. So software is getting simpler for people to use. Software is getting lighter so it can be used on smaller and smaller devices. So mobile phones are particularly important nowadays. And we’re using those to collect data in a more efficient way.
Skip to 2 minutes and 47 seconds So if we’re in the field collecting information, instead of having to write down the village name, they can just take the GPS coordinate from their phone, it goes straight into the database, and that’s linked to the village. And therefore, we know all of that information is correct. We’re starting to use free phone numbers and look at how we can collect information remotely by allowing people to text in messages or phone in through a free phone number with simple things like case at a village location. Or that there’s a body that needs collecting in a certain location. The main challenges in working in an outbreak like this is that Ebola is particularly a rural disease.
Skip to 3 minutes and 26 seconds And that means a lot of the places we’re working are rural areas which have very limited mobile phone connectivity and almost no internet connectivity. And this means that we have to look at different ways that we can collect information. So one of the key developments that we’re looking to get around that is the use of simple SMS messages to transfer information. So instead of needing an internet connection, you can transfer it using a standard mobile phone low bandwidth. Creating software, which works on your phone but doesn’t have to be online so that you can collect all the information, and then when you come back to somewhere with an internet connection, it automatically gets uploaded.
Skip to 4 minutes and 8 seconds There may also be issues around confidentiality. Ebola is a very sensitive disease. So data which may not have been sensitive in another outbreak may be sensitive in this case. And we have to look very carefully about the data sets that we collect. Are we storing them in a sensitive manner? Are we making sure they’re secure? Are we being more aware of which data sets we’re making publicly available to ensure that there’s nothing that is made publicly available which either would allow cases to be identified or other information which may upset local communities.
Skip to 4 minutes and 40 seconds When we’re looking at what data different groups need, whether it’s a government or whether it’s someone working very locally, it’s surprising how often they need the same types of information. So the government will need the individual cases, although they may group that data to look at what is going on at a much larger level, while a local fieldworker may want to know specific locations to allow them to visit a location. But overall, they’re using very, very similar data sets. They’re requirements are very similar. What tends to change is the skills that they have, the access to the data they have, where they’re working.
Skip to 5 minutes and 16 seconds And that’s what we need to be aware of, making sure that the data is available in a format and that they’re supported to use the data as they require. One of the things we need to improve is on information management and the idea of what is included in this. It’s not just about the data sets we collect. It’s all about the way we communicate information, both between health professionals, the IT people, but also with communities. How communities can interact with the data, how we talk to the communities about what’s going on. One area around this is the information about social behaviour that we collect. We’re missing a lot of this at the moment.
Skip to 5 minutes and 52 seconds It’s not considered part of the information management. And if we are more aware of this when an outbreak happens, we may be better placed to collect that information as it goes on so we can see when behaviour change actually occurs.
Gathering and disseminating information
In an outbreak on this scale a key issue is to know what is happening where. This is not easy in areas where there are no detailed maps as well as incomplete data on the cases. In this video, Chris Grundy describes how information is being collected and collated on where cases are occurring to produce information that is useful for those working at the local level and those planning the response.
A certain amount of information is collected on each patient. This includes the village they come from and data on contacts. Areas are being mapped using satellite imagery and crowd sourcing data to be used with OpenStreetMap.1 Mobile phones are playing an important role. Freephone numbers are available for people to text in information on the location of cases or bodies that need to be collected. Health workers can record the GPS coordinates on their phones when they are in a village even where there is no phone coverage so location data can be recorded accurately. All datasets need to be treated carefully to maintain confidentiality.
Workers at different levels have different requirements in how they want to see information, although the basic data underlying it will be similar.
Maps are drawn to show the location of treatment centres. When overlaid with the actual locations of recent cases, which are not available publicly for confidentiality reasons, it can help with the planning of where future centres might be needed.
The problems of case definitions, missing cases, and collecting data in care centres are addressed in later steps.
© London School of Hygiene & Tropical Medicine