I am currently finalizing the development of an HIV Patient Tracker on DHIS2 version 2.40.5, inspired by the 2.0.1 package. I would like to know the best approach to calculate the total number of PLHIV being followed.
Additionally, I’ve attached a screenshot of an indicator configuration for this program. The indicator is named “HIV CS - PLHIV in the cohort: 0-4 years, Female.” Could you please help me better understand this specific setup and its purpose? Any insights or documentation on this would be very helpful.
thank you for your post and we are very glad to hear that the HIV toolkit has been used as inspiration for your implementation.
Related to the calculation of PLHIV being followed, first it’s necessary to have the correct definition of it for your implementation.
In the toolkit the one used and agreed with WHO was:
“All HIV positive patients that have been visited and/or have reported their status (alive, dead, LTFU) durign the reporting period”
As you can see, the analytics period boundaries uses a cutom boundary target in which case the Program Indicator will count only the patient that have a visit during the period: From the date of diagnosis (#{K5ac7u3V5bB.fxXDe8OZ86q}) to the last date of visit in which an HIV status have been reported (#{ang4CLldbIu.CrFaWOLSKiK}).
In the HIV documentation you can find the explanation under the Program Indicator section.
We know very well the challenge of having several HIV patients that should be “marked” as Lost To Follow Up (LTFU) but that get cumulated inflating the cohort with a direct impact on the program indicators.
To improve the quality of this indicator you can either:
Periodically doing a manual cleaning assigning the LTFU status: very time and resources consuming and difficult to maintain over time
Automatically exclude the LTFU: through the use of custom period boundaries and filters you can include only the ones that are not considered LTFU
How to do it? First of all you need to have a clear definition of the LTFU. According to WHO a patient should be considered LTFU when more than 28 days has passed since the last missed appointment or last ARV refil once the ARV diistributed is finished.
In this case the Program Indicators will have only one custom period boundaries (date of diagnosis) and in the filter you’ll need to include only the ones where in the last visit they were considered “alive” (not dead or LTFU) and that the difference between the analytics period end (V{analytics_period_end}) (you should include as well the V{current_date} in case the current date is included in the analytic period) and the next planned visit date (V{due_date}) + 28 days (baed on WHO definition but this need to be adapted according to the LTFU definition used in the implementation) is >0 OR in which the difference between the analytics period end (V{analytics_period_end}) and the last day with ARV is >0 as well.
Of course there are more factors that will need to take in account as for example the due date is not present than will need to refer to the last day with ARV, etc.
The Program Indicator " PLHIV currently on ART" use the same exclusion logic as described in the guide
I hope this answer doesn’t create more confusion than clearity! In case you want to have more detailed information don’t hesitate to contact me and we can plan a meeting! I’ll be delight to better understand the implementation!
Thank you very much for your detailed explanation and for sharing the relevant documentation. Your insights have been extremely helpful in refining the setup of our patient monitoring system for PLHIV.
I’m pleased to inform you that the configuration of the Program Indicator has been successfully implemented and is functioning as expected. The calculation of the number of PLHIV is now automated, taking into account the time boundaries between the HIV-positive diagnosis date and the “HIV - Cohort date.”
Looking forward to staying in touch and exchanging more insights in the future!
