Palindrome Data (Palindrome), a data science implementer that specialises in alternative data and machine learning tools for community development, has created South Africa’s first suite of digital and paper-based HIV tools backed by machine learning, designed to help frontline healthcare workers triage at-risk patients.
The solution, leveraging machine learning and multiple data sources, is designed to be used in both digital and paper-based environments so that healthcare workers can identify and manage high-risk patients and relevant interventions to increase HIV treatment retention and mitigate the risk of loss to follow-up (LTFU).
The solution can correctly predict a patient’s viral load (suppressed vs. unsuppressed) for three out of four patients; and can anticipate two out of three times when a patient will drop out of care.
“The biggest obstacle facing HIV patients is dealing with an overburdened healthcare system that can’t afford to take the time to deal with their unique challenges,” says Lucien De Voux, Director of Market Strategy at Palindrome Data. “There is a need to retain and engage patients in a relevant way. Our mission is to optimise the delivery of care between providers and patients, and we pursue this with a solution that anticipates when patients will thrive and when they will struggle.”
With one of the most significant HIV burdens in the world, South Africa is managing the largest ART program, but it is consistently struggling to apply those resources efficiently at scale. Clinicians and facilities are under pressure to do more, with less. The solution developed by Palindrome is designed to unlock much-needed capacity and drive a measurable impact on the efficacy of these ART programs by retaining patients in care and reducing the number of people able to transmit the virus.
“After years of research and development with healthcare partners and epidemiologists, we developed a suite of tools that leverage both paper and digital environments to help clinicians establish which patients are most at risk so that they can shift their focus to those patients to ensure retention and ultimately improve outcomes,” says De Voux.
The solution provides clinicians with essential insight into which patients are stable and which are most likely to return in addition to allowing them to shift their attention to high-risk patients. Knowing an adverse event is likely to occur before it happens can open up a new toolbox of interventions that can be deployed while the patient is sitting in front of the practitioner. The solution also underscores how the public healthcare sector can fully realise the untapped potential of its data and use this to transform patient care on a national level.
“We combine new and alternative data sources along with traditional data sources to generate relevant insights into behaviour and patterns in developing markets,” says De Voux. “Our goals are to adapt and tailor advancements made by the private sector to have an outsized impact on humanitarian initiatives; and to work with the public sector to create solutions that help them maximise their resources and transform their systems.”
Palindrome, with its partners, have developed several papers that unpack the data models used to develop the solution, along with the relevant statistics and insights that highlight the value of the solution within the market and with respect to patient care.
● HE2RO Policy Brief: PREDICTING SUPPRESSION OF NEXT HIV VIRAL LOAD IN SOUTH AFRICA
● VL & LTFU Pre-print_Feb21.pdf: latest publication preprint on how we use machine learning to predict retention and viral suppression
● Determining Vertical Transmission routes of HIV exposed infants
