Encouraging health adherence

A story inspired by a personal health scare helped a team of medical students win the 2026 Health Systems Hackathon. The hackathon was recently hosted by the Faculty of Health Sciences’ (FHS) Health Systems Innovation Hub, in collaboration with Harvard Health Systems Innovation Lab (HSIL).

The University of Cape Town’s (UCT) Ismaeel Noor Mahomed, Kevaal Govender, Timilehin Aderibigbe and Ashraf Moosa are playing their part in ensuring adherence to medication for patients with long-term conditions. It was Moosa, who, last year, found himself diagnosed with tuberculosis (TB).

“I completed treatment because I understood the diagnosis, why adherence mattered, and what could happen if I stopped treatment early. That made me realise how much of a privilege health literacy can be,” said Moosa.

Their patient-facing mobile app, Siyaphila, stands out for its integrated artificial intelligence (AI) and advanced analytics capabilities. As patients use the app, the platform captures structured adherence data over time. This data can then be used to identify early warning signs of non-adherence, flag patients at risk of complications, and support more personalised, preventive care. In the long term, the goal is to use this data to train predictive models that help clinicians, funders, and medical aid companies intervene earlier – before complications and unnecessary costs escalate.

“One of the most interesting things for me was how non-adherence is rarely caused by just one issue. It is often a combination of poor understanding, side effects, access barriers, cost, and weak follow-up systems. It also became very clear that the cost of supporting adherence early is far lower than the cost of dealing with complications later,” Moosa added.

“There will be a lot of benefit for healthcare givers and patients with this app.”

Fellow team member Govender said he was confident of their chances of winning at the two-day event. “We came up with a good pitch.” The outcomes of the hackathon reflect a broader goal: answering the call to position UCT as a leading African university for innovation-driven, inclusive development and entrepreneurship, as outlined in Strategy 2030.

As hospital staff interns, Aderibigbe and Moosa bring a valuable advantage: firsthand insight into the factors that often lead to patient admissions. “There will be a lot of benefit for healthcare givers and patients with this app. I think for the healthcare system there is a huge financial benefit, in that with reminders, you reduce the risk of health complications from non-adherence. Complications are more expensive to treat,” said Aderibigbe.

There is still a long journey ahead for the students, including participation in bootcamps. “I think because of our team, we may lack a bit on the technology side in terms of needed support going forward; however, at the hackathon we already started a relationship with Techzarlogy’s Llewellyn Petersen who has offered to assist us in creating a prototype and so we can continue carrying on into the next phase,” Mahomed said.

Personal data

Good health and well-being are urgent Sustainable Development Goals (SDGs) imperatives, and Mahomed feels the built-in features they plan for the app (such as educational content) assist in matching adherence with information sheets for patients to know what medicine they are taking and why. Moosa expanded on this: “I think this intervention contributes by supporting continuity of care, improving patient understanding, encouraging better long-term adherence, and potentially reducing preventable complications. It shifts care in a more preventive and patient-centred direction, which is essential if we want health systems to improve outcomes in a sustainable way.”

Govender added: “We focused on offering small, bite-sized pieces in a language they can choose. We want to make it as user-friendly as possible. We used TB and HIV in South Africa as our starting point, but it very much is focused on a global point of view. We want to train the AI model to use predictive analytics to monitor who is at risk, why and intervene early.”

“The long-term vision is to move from reactive care to more personalised, preventive care.”

The integrity of their project also lies in how they will protect people’s personal data. Aderibigbe said: “I did my third-year elective on health data, and I think the context we are in, I find that health data is not used in a way that it should in terms of predictive analysis and improving the healthcare system.”

Moosa said: “Any real implementation would need strong privacy and governance measures, including informed consent, secure storage, access controls, and appropriate anonymisation or de-identification where data is used for analytics. This is something I see as essential for the idea to be implemented responsibly.”

Long-term vision

He concluded: “One important aspect of the idea is that it was designed to be scalable. While we presented it as an app-based platform for chronic disease adherence, we also thought about how a lower-cost WhatsApp-based version could be used in more resource-constrained settings. More broadly, the long-term vision is to move from reactive care to more personalised, preventive care by using adherence data more intelligently.”

*Ashraf Moosa was not available for photos; however, he did participate in the interview.