In business, when reflecting on historically famous projects, such as the Apollo program or the Manhattan project, the stories often focus on the genius of individuals. But genius by itself didn’t, and couldn’t, put astronauts on the moon. The key is collaboration and co-ordination. However, the efforts of nutritionists, programmers or nurses, for example, are rarely recognized when recounting these particular tales.
There are many parallels between how we think about these projects and the rise of data science and analytics groups’ inside businesses. These groups are shedding light on the art of the possible, but without the right design, collaboration and delivery mechanisms, this ‘possible’ isn’t deliverable. This is a particular problem in population health initiatives – only made more prevalent by the COVID-19 pandemic.
Many healthcare organizations see data scientists and analytics groups as the key to seizing the opportunity afforded by fundamental shifts in the healthcare landscape, such as risk sharing between payers and providers, and regulation on interoperability. But while these capabilities can light the way, these groups tend to sit within their own siloes and lack a view of what would make their work easily actionable. Harvard Business Review reports that: “Fragmented care is an important part of the reason for high costs and utilization… Patients receive care through a patchwork of providers at various sites – outpatient dialysis units, primary care practices, specialty clinics, hospitals and others – which often don’t communicate. Gaps in care are inevitable, and opportunities to intervene before problems arise are often missed.”
Opportunities missed are just future opportunities. Organizing healthcare will deeply impact individuals and communities and being part of such a positive change will elevate the efforts and accountability of people turning insights into action.
So, how can healthcare organizations capitalize on insights and use them to drive deliverable population health initiatives?
Developing insights within a data team and sharing them with other parts of the business won’t deliver action. It’s critical to bring together a cross-functional team of data scientists, clinicians, care program operators, and other subject matter experts to generate relevant, contextualized insights. Such diverse teams uncover previously unidentified value streams, share accountability, and quickly assess the feasibility of action. Leaders can then prioritize innovative ideas based on their projected ability to improve health outcomes, the organization’s ability to influence them, and the size of the potential return.
Keeping this cross-functional team together through testing the hypothesis will then help the analytics groups better understand the gap between measuring an opportunity and delivering on it.
As an example, the cost savings from 100 per cent adherence to medicine reconciliation is staggering, and easily measured from claims data, with annual costs of non-adherence ranging from $100bn-$290bn in the U.S. alone, according to the New England Healthcare Institute. Yet the hurdles to intervention are hard to see without the input of downstream stakeholders who understand the financial restraints on a primary care practice as well as patient behavior. Participation from all stakeholders touched by a value stream not only illustrates the complexity of implementation but, more importantly, brings a sense of purpose and accountability to this team.
Value from this participation is summarized by Theodis Ray, a janitor at Kennedy Space Center during the Apollo 11 mission who went on to be the logistics lead for United Launch Alliance: “It changed my life because it made me a better person than who I am… This is where I got a chance to stand my ground. You had to be firm because they challenge you.” Everyone has tremendous value they can bring to a project.
To turn population health insights into care interventions and actions, healthcare organizations should use key concepts of Agile, like sprints, Scrum teams, and minimum viable product. Such techniques reduce the risk of financial loss, facilitate collaboration, and accelerate time to value.
For example, a traditional approach causes many organizations to spend substantial time and money pursuing ideas that don’t drive value, or no longer drive value by the time the idea reaches market. Operating in sprints delivers value incrementally and creates natural break points. That makes it possible to assess whether ideas are delivering as expected and to quickly deprioritize those that aren’t.
While the genius behind the Apollo program lit the way, it could never land an astronaut on the moon by itself. The engagement of programmers, nurses, and janitors made genius a reality. Each brought a piece of the puzzle and together formed a team with greater purpose and accountability. This is also how brilliant insights from data science teams can drive action in healthcare organizations. The data science team shows how others can use their skills, from testing and quantifying the hypothesis, through designing the intervention, to measuring the results.