Realizing value from data and AI
Data is everywhere
An increasing number of devices are collecting health information, and more people are staying connected by tracking their own health data using technologies like smartwatches, fitness trackers, wearable monitors, and sleep monitors—to name a few. Information collected in healthcare settings continues to be fragmented due to care delivery happening in a variety of places, from primary care offices to pharmacies and labs. This causes information and data to be stored in disparate systems, making it challenging to collate information for a holistic patient view.
Despite the challenges, collecting and managing a person’s healthcare data is critical to ensure individuals get the right care, when and where they need it. Value-based care, accountable care organizations, and payviders continue to push for more holistic care of the individual.
There is also a need to understand social determinants of health at an individual level, and collect this data to ensure patients get the proper support and interventions. In addition, this data can be used to help with disease prevention and prediction.
Four pillars for developing a holistic data strategy in healthcare
Healthcare organizations need to continue modernizing their technology systems to fully use data to support innovation and drive and facilitate data exchange. However, simply adding technology won’t create change; it’s critical to develop a data strategy, ensuring governance prioritizes and makes data relevant, create a robust operating model, and invest in a modern technology platform.
1. Data strategy and strategic alignment
Not only is it important to have a vision and strategy for data within your organization, but there also needs to be alignment across leaders to foster a data-driven culture.
Evolving culture to ensure the organization understands the importance of data and how to use it to drive specific outcomes requires thoughtful change management. Too often, we see organizations deprioritize the training, tools, and reports that help build data acumen, measure results, and drive clinical outcomes. By ensuring their people are sufficiently supported, organizations are more likely to drive adoption and long-term success across the myriad of stakeholders within their healthcare ecosystem.
A strong data strategy needs to incorporate partnerships and address robust data sharing and exchange agreements to overcome the fragmentation of data storage.
2. Governance
Governance is needed to ensure investments in the right initiatives, establish alignment across the organization, manage prioritization and transparency of data initiatives, and meet the overall clinical and administrative objectives. Having the right governance structure requires an executive steering committee and a program governance committee to make decisions and address risks, issues, and conflicts.
In addition, a focus on data governance ensures the data is trustworthy, of high quality, secure, and has clinical and operational data ownership defined so that data platforms aren’t creating a garbage-in, garbage-out scenario.
3. A robust operating model
Often, we find organizations that struggle to deliver data initiatives lack a robust operating model. Having one in place helps bridge strategy and delivery by outlining the high-level data capabilities and associated ownership, roles, and responsibilities. It should also address the need for a cross-functional team within and outside of IT, along with ways of working and cultural norms for delivery.
An agile/scrum methodology can help ensure speed and reinforce regular delivery check-ins with clinical and operational users. When onboarding new technologies, leadership should evaluate the talent and skills of their teams alongside organizational design and reporting structures for alignment. Lastly, organizations should invest in processes to enable high-performing teams to execute efficiently.
4. Components of a modern data platform
Modern data approaches an ELT versus ETL process (i.e., extract, load, and then transform the data versus extract, transform, and load). From data ingestion and processing to consumption and analytics, organizations should focus on both modern tools and the team training and support necessary to thrive. Considering all end users is critical, so both clinical and operational teams are served and supported by your approach. The right combination of technology and human investment can accelerate processes and lead to richer, more valuable insights.
Your modern data strategy and platform should address/include:
Data ingestion and processing
A data lake or ocean to store raw data (often used to support AI and GenAI initiatives)
A data warehouse for curated data modeled for analytics and reporting
Orchestration, logging, monitoring, and deployment
Data governance tools
Consumption and analytics
Training investments
Leveraging GenAI and new technology to improve healthcare delivery
Organizations must have a modern data platform in place to capitalize on the boundless possibilities of GenAI. From improving care delivery and navigation to increasing productivity and efficiency to disrupting the industry with novel and pioneering solutions, healthcare is on course to be more efficient, personalized, and innovative in 2024. We are working with our healthcare clients to support this innovation and have highlighted several use cases here.
GenAI and the continuum of care
While AI and GenAI are all the buzz, many are skeptical: Will it meet the hype? At Slalom, we are seeing evidence that it will and predictions that 2024 will be a year of adoption and increased use. We are already seeing healthcare vendors and tools incorporate GenAI into their applications as well as partnerships to support the acceleration of AI-embedded workflows.
When considering possible AI use cases, healthcare organizations must prioritize responsible applications of AI. The appropriate governance and training must be applied to account for potential bias and/or hallucinations. Ultimately, it will be up to players across the healthcare ecosystem to focus on responsible applications of AI totake accountability for AI and support improved patient outcomes. We are optimistic that, with a mature vision, clear roadmap, proven proof of concept, and supported teams, organizations will be enabled to deliver on their promise to patients—at scale.
Slalom contributor: Brenda Yoo-Young
