HIVHACK.org - Open #data4good Hackathon page
HIV Drug Resistance data4good Hackathon.
Brussels, November 23rd & 24th, 2018
HIV Drug Resistance Challenge
In the context of ART roll-out in low- and middle-income countries (LMIC) around the globe, and the recommendation in the most recent WHO guidelines to apply the “Test & Treat” (T&T) strategy, over 17 million HIV-infected persons have received ART worldwide. The monitoring of those persons is still suboptimal; they are maintained on a failing treatment regimen for long periods of time. This results in the accumulation of drug resistance mutations, and in immunological deterioration. Both these factors contribute to a poorer efficacy of 2nd line regimens for those patients.
High levels of pre-therapy drug resistance are observed in the WHO HIVDR 2017 report(1). If, in sub-Saharan Africa, levels of NNRT Pre-Treatment Drug Resistance exceed 10%, if NNRTI in first-line ART, between 2016 and 2020, Pre-Treatment Drug Resistance will be responsible for(1) 105,000 new HIV infections, 135,000 AIDS deaths and US$ 650 million in ART drug cost.
Through its Global Public Health organization (GPH), Johnson & Johnson (J&J) aims to deliver integrated evidence-based solutions to address complex global health problems in mental health, HIV/AIDS, tuberculosis, soil transmitted helminths and vaccines.
GPH Disease Management Programs (GPH-DMP) is conducting a collaborative exercise, involving stakeholders from various disciplines, to develop a model to map HIV drug resistance hotspots in LMIC’s.
Through this exercise we will get an understanding of epidemiological and non-epi factors influencing the emergence of HIV-DR and a model mapping the spots. This will result in dynamic and up-to-date graphical and visual representations of HIV-DR maps allowing governments, health workers and other stakeholders to be informed on the topic and to turn the understanding of the drivers into actionable items in the prevention and countering of HIV-DR. As part of our mission we intend to educate on the approach and outcomes locally.
To date, GPH-DMP has implemented a model globally mapping the probability of observing HIV-DR above a 10% threshold as determined by WHO(1). GPH-DMP is working with BlueSquare HUB, a Belgium based company working with a variety of governments, NGOs and donors to reach over 100 million people in 18 countries. BlueSquare is developing an interactive dashboard visualizing HIV-DR related risk indicators in the Democratic Republic of Congo.
J&J GPH is also sponsoring a “Data for Good” event, organized by the European Data Innovation hub, a non for profit organisation grouping the major european datascience communities based at DigitYser (digityser.org) in Brussels.
It is our long-term goal to reach out to more countries and stakeholders, further refine the model and continue to help and inform countries, governments and organizations on HIV-DR.
In preparation of the Data4Good Initiative, some data already has been gathered from publicly available websites in countries such as Tanzania. We are asking for your help to complete our “Data for Good” data set(s) on HIV Drug Resistance and for additional contacts in this field.
In particular we are looking for Viral Load Suppression and/or HIV-DR data aggregated at facility or regional level over a reasonable time span, thereby avoiding the use and/or disclosure of any information relating to an identified or identifiable natural person. We are also calling for influencing factor data such as, but not limited to, prevalence, ARTcoverage, education level, healthcare spending at the same spatial and temporal granularity.
The model will be developed under MIT license. At the request of the contributor, location-based anonymisation will be applied to the data used to build and validate the model. Data ownership will remain in any case at the contributing party. If required by the contributor, individuals using the data will be under NDA, restricting its use to the objective of the “Data for Good” initiative and committing to destroy the data after the event. Datascience techniques will be used to enrich the original datasets to visualize.
Under MIT, contributing parties will be granted to use the model, free of charge, provided that all copies of the licensed software include a copy of the MIT License terms and the copyright notice.
The contributing party shall receive the benefit of:
Global Public Health
Johnson & Johnson Global Public Health
Philippe Van Impe
Founder & CEO
(1) Source: WHO HIVDR 2017 report https://www.who.int/hiv/pub/drugresistance/hivdr-report-2017/en/