Healthcare AI 2024
Use Cases Expanding to Meet New Market Needs

Partnership & Expertise from ClosedLoop Drive Consistent Outcomes for Health Plans, ACOs; N1 Health’s* Collaboration & SDOH Algorithms Achieve Results for Health Plans

Interviewed customers of ClosedLoop—many of whom are ACOs and health plans—describe the healthcare-focused vendor as collaborative. They say ClosedLoop provides strong account managers who partner with customers to deliver initial and ongoing guidance, training, and expertise. Customers are paired with a data science specialist to help achieve adoption and outcomes. 50% of respondents have expanded the use cases for which they use the solution. Reported outcomes include identification of high-risk members for care management, reduction in hospital admissions, reduced costs due to fewer negative outcomes, increased drug adherence, and increased screening measures. The solution brings in SDOH data, and many customers highlight its excellent integration with both payer and provider systems. A few customers would like data from the system to be better integrated into EHR workflows. Customers are optimistic about future development, including an improved user interface.

N1 Health* respondents include payers and provider-owned health plans. The technology improves member engagement and risk capture by using SDOH algorithms to better target members. Most respondents mention strong partnership and feel the vendor is responsive and collaborative. Reported outcomes include higher-than-expected outreach numbers, improved risk capture, improved marketing campaigns, improved member engagement, improved Medicaid reimbursement, and more-creative strategies for approaching underprivileged populations. Two respondents report that outcomes have not been realized yet, due to being early in the process or to not having yet acted on the data. Additionally, some customers note that the growing customer base sometimes makes it difficult to get the support they need.

*Limited data

Epic Customers Expanding beyond Clinical Use Cases; Outcomes Lagging from Oracle Health* Due to Decline in Support

Epic’s and Oracle Health’s AI customer bases are comprised mostly of large health systems; most customers leverage the vendors’ prebuilt models rather than building custom models themselves. Epic customers have begun expanding outside their initial focus on clinical use cases to adopt operational and financial areas, and they report success with those models. As the information is embedded in the EHR workflow, customers report good adoption from employees. Reported outcomes include clinical documentation improvement, improved quality measures and patient outcomes, substance abuse prediction, improved denials management, identification of high-risk pediatric patients, reduced readmission risk, improved sepsis prediction, reduced no-shows, prioritization of work, objective and consistent communication of severity of illness, and improved risk capture. Customers report that previous gaps in support have been improved. They describe Epic as helpful and responsive and say they have a dedicated support person. Improvements to the documentation and training have made the solution easier to implement. Approximately one-fifth of respondents continue to report frustration with nickel-and-diming due to how Epic prices the models. Several customers also report that integrating outside data into EHR workflows is a challenge.

Feedback from Oracle Health* respondents is split. Regardless of their overall satisfaction, customers often report that it takes a significant internal lift to get the solution to work how they want. Outcomes reported by the more-satisfied customers typically center on clinical data and include readmission prevention (including for COPD patients), creation of disease-specific registries for research, improved stratification of patient interventions, and decreased patient falls. These customers also often mention optimism about Oracle Health’s plans to move to the cloud. Dissatisfied customers say support has continued to decline post-acquisition and feel they are left to their own devices to make the solution work. About half also feel nickel-and-dimed, with many saying this has increased since the acquisition.

*Limited data

Outcomes with Health Catalyst* Not What They Could Be Due to Gaps in Usability & Training

Other Key Market Trends

Data science platforms are most commonly used by larger organizations (500+ beds).

Most interviewed organizations leverage data scientists to use and monitor the solutions.

Provider organizations most commonly use AI technology from their EHR vendor, supplementing as needed with third-party vendors or homegrown solutions.

Most organizations using their EHR vendor’s data science platform use prebuilt models rather than building their own.

Importance of workflow integration: Many respondents note the importance of training and ensuring the data is actionable and embedded within user workflows. Organizations continue to want guidance and partnership from their vendors to help them develop an AI strategy and encourage adoption. High-performing vendors—such as ClosedLoop, Epic, and N1 Health—tend to be rated higher in these areas than other vendors.

Expansion of use cases: 50% of respondents report expanding into additional use cases, though the majority of outcomes for both health plans and provider organizations still relate to population health or finding high-risk patients. Adoption of operational and financial use cases is increasing as organizations look for ways to alleviate the staffing and financial challenges many health systems are facing.