OHDSI Home | Forums | Wiki | Github

Collaborators/data partners: HIV Antiretroviral Treatment Patterns Study

Hello all,

We are interested in collaborating and working with data partners on a new study that has been posted to the OHDSI Research network: (Study Repo)

This purpose of this study is to study characterize and evaluate trends in pathways for antiretroviral therapy (ART) for individuals who have been diagnosed with Human Immunodeficiency Virus (HIV) and treated persistently over two-years. Both single-ingredient antiretrovirals and co-formulated combinations will be evaluated. The analysis package has been developed using the CohortPathways and Characterization tools in ATLAS and will be performed in the ATLAS environment.

Established as an essential public health intervention, ART has been extremely important for the achievement of viral suppression to both improve HIV-related morbidity and mortality and to prevent transmission. Antiretroviral therapy involves taking a combination of medications daily. Regimens can be complex and are often dependent on a variety of clinical and personal factors. Observational research provides a unique opportunity to evaluate heterogeneity in treatment for HIV and provide insight into real-world patterns in ART. There are somewhat few network studies within OHDSI that assess complexities related to care for microbial diseases and to the best of our knowledge, real-world evaluation of HIV antiretroviral therapy across a data network has not yet been conducted. We seek to reveal patterns in antiretroviral treatment for individuals diagnosed with HIV through identifying commonly used first-line antiretrovirals, evaluating temporal trends in treatment, and comparing antiretroviral treatment in the real-world to clinical recommendations.

A draft of the protocol is available here: (Protocol Draft)

Please reach out (raechel.davis@yale.edu) if you are interested in collaborating or participating!


This is tremendous, thanks @redavis for posting this. It’s really wonderful to see an Early Stage Researcher taking the initiative to lead a global network study in an important area. I’m keen to participate and really hope others in the community will join as well!

This is particularly a great opportunity for those data partners who have ATLAS installed and are looking for a light-weight way to participate (simply import a JSON spec and run!) and contribute results to a study that will almost certainly result in interesting findings and a nice publication.

Let’s see what HIV drug utilization looks like across the world!

1 Like

Thank you so much for your support Patrick!

Big fan of the topic/study. Currently I can’t provide any data as site. :frowning:
In our past analyses (Medicaid dispensation data), HIV Ryan White and other assistance programs can impact what data is available in claims.

Is that somewhere published, @Vojtech_Huser?

1 Like

Very excited to see this study, @redavis!! :grinning_face_with_smiling_eyes: Happy to run this on our PharMetrics Plus data at Northeastern.

@Chris_Knoll / @anthonysena , Raechel and I have been talking about ways to make those sunburst plots with the downloaded outputs from the Treatment Pathways tab.(Maybe there is a way to export them now and I’m behind the times… this is also plausible.) Do you have any guidance on how she could code that? My brain is super rusty on this. I’ve lost the helpful pointers that my ol’ pal, Henry Morgan Stewart, gave me back in the day. :woman_facepalming:

1 Like

Vojta, thank you for your support as well as bringing this up. I would definitely like to discuss what you have observed in past analyses if you have any further information!

@krfeeney - Atlas v2.12.0 introduced the ability to export/download visualizations. You can see it working here: ATLAS.

When you hover over the visualization, in the top right corner you will see the option to export as a PNG/SVG:

Hope this helps!


Solid! I’ll have to take a look at the power of ATLAS v2.12. I’m not sure if all of our collaborators have moved forward and realize this benefit. Proof even old dogs learn new tricks.

1 Like