COVID-19 Surveillance Webinar - Q&A and Discussion Thread

Any COVID Predictive modelling or algorithms implemented so far. The key to the fight against COVID is having inbuilt capabilities to be able to predict regions likely to be affected by the pandemic. Android Apps and dashboards could then be used to respond with interventions e.g deploy preventive teams, testing teams etc.
Basically my question is how do we incorporate some predictive algorithms within our DHIS2 Platforms. Are there tools (use case) examples available for integration that may have worked for other outbreaks?

Is there a way to generat the covid19-surveillance-report with the data collected from tracker instead of enter them again manually in the data Entry app?

1 Like

On the android version, do we have the “Register” or “New case” on COVID-19 Case surveillance just for registering a new case using android not only seeing the enrolled cases registered using web version?

Best,
Josue

Hi Gysanjaya, some DE from the Tracker are reused in the Event program. Please, specify your question, so I can provide more detail.

1 Like

Hey Derebe, this part is a bug in the Capture application. Here is a jira issue: https://jira.dhis2.org/browse/DHIS2-8519
The issue is fixed, and will be reaching you in the upcoming DHIS2 point releases.

1 Like

You could use a Client generated UID.
This UID would uniquely identify an individual through out the program stages.

1 Like

If you start using tracker and move to an event system when the cases increase, does the data integrate seamlessly? i.e. Will the number of suspected or confirmed cases recorded in the tracker module link directly with the events module? i.e.do they share the same data element or have separate data elements been created for each program/module?

1 Like

OK - I just see your response:

“Hi Gysanjaya, some DE from the Tracker are reused in the Event program. Please, specify your question, so I can provide more detail”

So I assume it does??

1 Like

Hi Sarah,
As Andrew has linked, the supported DHIS2 versions will always be specified in the release documentation. In cases where the package is supported in more than one version of DHIS2 (i.e. aggregate package) we include a separate metadata file for each DHIS2 version. The current recommendation is to use a 2.33 instance for the tracker packages as this will ensure compatibility and the use of latest functionality.

1 Like

Hi @enyankesha
The localization of contacts could be captured by the TEI coordinates. These can be displayed in a map. You can read more about this feature in the [DHIS2 Documentation]. (Home - DHIS2 Documentation)

1 Like

Hi @Gysanjaya and @mbuttsworth -
The Case-based surveillance tracker package captures the most complete set of data elements and is aligned with the WHO data dictionary for case-based reporting. The event program is designed with a small subset of these data points. It is intended for use when caseloads become too high and case-based reporting is not feasible in a timely manner (a learning from Ebola outbreak surveillance experiences). You can read more about the use case and data elements for the Surveillance Event Program in the COVID-19 Surveillance Event Program System Design Guide (v0.1)

1 Like

Hi @muhireandrew2020 this is indeed still a problem. We are unable to create relationships across different organization units/geography. This is coming in 2.34 but it is not yet possible.

1 Like

Hi @Gerard_Bisama, the event program is meant for when the system gets overwhelmed and you are unable to accurately process all of your cases in tracker. It is not meant for case management, but for tracking the epidemiology of the outbreak

1 Like

Hi @Josue the registered cases on Android get synchronized back to the web as long as there is internet available.

1 Like

Hi @Morley there are certainly mechanism to implement these types of algorithms into DHIS2. We would be interested to hear more detail of your thoughts on this.

We do not have examples for COVID-19 as this is our first time dealing with this virus, but we would be very interested to learn about required modeling for thresholds if available.

1 Like

Hi @Gerard_Bisama

Yes, all of this data can be generated directly from tracker

1 Like

Hi Michael,
If a country starts with tracker then moves to event, the indicators can be updated to combine data elements from the event and the tracker program. As currently designed, the event program package comes with a set of indicators based on data elements only from the event; and the tracker package comes with a set of indicators generated only from data elements in the tracker. Combined indicators are configurable through the user interface, it does not require a programmer. It is important for the country to make a clear implementation decision on switching from tracker to event to avoid any risk of double counting.

1 Like

Hi all! The recording from this webinar is now available on the DHIS2 YouTube channel: DHIS2 COVID-19 Surveillance Webinar - YouTube

1 Like

Just FYI @Shurajit_Dutta & @muhireandrew2020, this was also backported to 2.33 and included into the recent 2.33.3 patch.

A post was split to a new topic: Calculating contact tracing indicators in Pivot Tables