Applied Epi Community

Is open source software the path forward for surveillance?

I’ve been working with data from several countries across the European region and struggle with cross country analyses due to differences in data collection and reporting. I’m wondering if open source software could offer a solution to this issue by standardizing, validating, and summarizing completeness of data.

Im not sure of the best way to set this up, and I suspect there are already many tools out there that can be used. Does anyone have any insight on how this could be a reality? Are there organizations out there attempting to do this? What are the major challenges involved?

Hope this stimulates some fruitful discussion, thanks everyone!

Interesting question and discussion to start.
In my case it has also been hard to standardize data collection from different projects as different contexts end up having different needs.
Maybe one way is to define the minimum data elements needed for reporting (have have same naming) and leave anything specific to each context for them to decide.
Once that’s defined, there exists tools that can assist in cleaning, validating, summarizing and reporting data. I am using R (open source) and it can help do that.
Happy to engage further

hey @agimm - super relevant discussion point!
I think probably the most effective way forward is to promote the use of data dictionaries and then have standard tags for common epidemiological variables.
UN OCHA has the hxl format - I don’t love the adding of a second row, nor the hashes, but if in a dictionary could work. I think if as a community we push for some additions to this then it could become really functional - because analysis scripts could then be recycled for loads of different datasets that have appropriate tags.
See this brief discussion here for more thoughts.

A standard but flexible platform would be useful as well. We’re currently exploring DHIS2 for this purpose, but we lack local experience and expertise in this platform specifically.

1 Like