Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The author is a senior fellow at the Centre for International Governance Innovation and Director, Artificial Intelligence and Innovation marketing, at Microsoft.
As digital technologies become widespread in our lives, they have enormous potential to influence health and wellbeing. To get the most out of them, we need a new form of innovation focused on those who face the toughest barriers. And that means developing these technologies with historically marginalised people rather than for them.
History shows that the implications of inventions typically emerge well after their initial adoption. Over time, we begin to see the possibilities: mobile phones with touchscreens; an internet that delivers critical information and services and connects people around the world; cars that can operate autonomously.
But the inventors cannot always foresee the effects of their inventions, nor whom they may leave behind. New technologies are often expensive, at least at first. They require infrastructure such as electricity, networks and roads. They replicate and, in some cases, amplify inequalities within the societies from which they emerge.
Regulations follow later: seat belts in cars, privacy protections for internet users, moratoriums on facial recognition technology, and so on. While technology regulation is essential to the public good, it is intended to protect people from measurable damage rather than foresee or prevent it in the first place.
We need to broaden our approach to responsible innovation to encompass methods that are inclusive by default and empower all people.
This not only means expanding technologists’ understanding of the implications of what they create. It also means co-creating with rather than building for marginalised communities and adapting existing methodologies — or developing new ones — to reflect the needs, voices and expertise of people who are often excluded from the innovation process.
What would an empowerment model for healthcare innovation look like? Here are three recommendations.
First, enfranchise patients, community organisations and vulnerable groups. The Covid-19 pandemic has made it clear that healthcare systems are often ill-equipped to serve the people most in need of care. Community health workers and organisations create a bridge between healthcare systems and marginalised community members.
There is an urgent need to support relationships between people in need of help, health workers, partners, organisations and funders — and ease the administrative burden so the focus can be on directly caring for patients. My colleague Mary L Gray founded Project Resolve with the Healthy Community Hub — a coalition of black and Latino community-based organisations in North Carolina — to do this.
Project Resolve makes use of open-source software and systems to support two approaches: Case-centric care focuses on services to each individual, such as helping someone with housing instability; while service-centric care offers a specific intervention — such as a vaccine clinic — to a group of people.
Second, adopt multidisciplinary approaches. Healthcare is highly dependent on data: lab results, patient histories and other factors critical to understanding individual and population health. But they often lack important context, such as people’s access to fresh food, insurance, transport or stable housing. These factors can determine a person’s ability to afford medicine, follow treatment or gain access to care in the first place.
The more we build digital systems to inform the future of healthcare, the more important it is to include wider expertise from groups such as anthropologists, economists, linguists, medical, security and privacy experts, data scientists, engineers and community organisers who deeply understand the factors that can determine good health outcomes.
Third, complement “big” data with “small” data. One of the biggest challenges in healthcare innovation is “small” or “sparse” data — data that are often lost, overlooked or stored on multiple, disconnected spreadsheets. Examples include patient histories scattered among community health facilities, urgent care centres, pharmacies and hospital emergency departments.
New privacy-preserving methodologies are critical to collecting, storing and learning from “small” data responsibly, and combining them with large-scale data to build a more inclusive and accurate evidence base.
Monitoring environmental pollution is one example where both big and small data have a role to play. Satellite imagery can provide a wealth of data on air quality at a planetary level, but it cannot tell us much about conditions and trends in a specific neighbourhood. Project Eclipse, from my Microsoft colleagues, includes locally sourced data gathered in collaboration with neighbourhood conservation partners across Chicago. The goal is to give these groups the ability to monitor the environmental conditions that directly affect their communities.
This approach of empowered innovation represents a step-change to the way we conduct scientific research and develop technology. But, most importantly, it requires the humility to value and learn from the lived experience of the people it is meant to serve.