Data Utilization, Investment Lagging Among Hospitals and Health Systems

A survey conducted by the Healthcare Information and Management Systems Society (HIMSS) and healthcare data analytics company Arcadia about the current state of healthcare analytics platforms demonstrates that less than 60 percent of health systems’ data is used to inform intelligent business decisions, but stakeholders underscore the importance of data access.

The report, which polled 55 healthcare executives, clinical leaders, and health information technology (IT) stakeholders, aimed to better understand data utilization within hospitals and health systems.

Respondents were asked questions regarding satisfaction with their organizations’ current analytics platform solution(s), integration of applications and technologies with data analytics platforms, challenges the organization has encountered with investments in analytics platforms or upgrades, and the reliability and use of analytics data.

Survey data show that access to high-quality data, organization size, analytics investment challenges, data accuracy, and data utilization play major roles in the current healthcare data analytics landscape.

Approximately 93 percent of respondents reported that quality data across workflows and platforms is crucial to a healthcare organization’s performance. Whether respondents agreed that this is the case was related to which electronic health record (EHR) platform their organizations primarily use.

Organizations that do not use Epic as their primary EHR reported higher agreement with the statement “Having access to quality data and the ability to utilize that data across all our platforms/workflows is critical to our organization’s system-wide performance.”

Amongst organizations that primarily use Cerner or another EHR platform, 100 percent of respondents agreed with the statement, compared to 87 percent of respondents whose organizations primarily use Epic.

Organization size also strongly influences analytics priorities. The survey indicates that larger organizations, defined as those with 15,000 or more employees, are more likely than their smaller counterparts to implement clinical decision support (CDS) tools within the next 12 months.

Of the organizations represented in the survey, 94 percent of larger organizations reported that they either have already implemented or plan to implement CDS tools within the next year, compared to 68 percent of smaller organizations.

Organizations with 7,500 or more employees were more likely to have plans to implement decision-making tools for their provider networks, with 38 percent of these respondents planning to implement one of these tools in the next 12 months, compared to 10 percent of organizations with less than 7,500 employees.

Within both larger and smaller organizations, 30 percent of respondents in C-suite roles indicated that they had plans to prioritize artificial intelligence (AI) and machine learning (ML) in the next year.

Across healthcare organizations, investing in analytics is a common challenge. Approximately 82 percent of respondents reported being satisfied with their current analytics platform, but 71 percent indicated that other strategic priorities create barriers to analytics platform investment. Limited staff resources are also a challenge in this area for 58 percent of respondents.

In terms of data accuracy, roughly a third of respondents reported that their data is less than 76 percent accurate. However, reports of data accuracy varied depending on the respondent’s role in the organization. Those in executive leadership roles reported a significantly higher average percentage of data accuracy compared to those in clinical or IT roles.

Data utilization for business decision-making also varied. On average, only 57 percent of an organization’s data was found to be used to make intelligent business decisions. Respondents from larger organizations were more likely to report that their data is being utilized in decision-making.

Those who work within integrated delivery networks (IDNs) or multi-hospitals were also more likely to report their data is used to make intelligent business decisions compared to their counterparts at stand-alone hospitals, specialty facilities, and academic medical centers (AMCs).

Similar research shows that the implementation of AI tools is also varied among healthcare organizations.

Research conducted by the Center for Connected Medicine and KLAS showed that two-thirds of respondents are using AI within their organizations, and AI use was most common among facilities with over 1,001 beds.

Pre-built ML models were most popular, with a majority of organizations of all sizes using these solutions. Applications of AI included health/disease management and prediction, with high priority placed on operational optimization investments by respondents in the next two years.