Leveraging Health Care Data Analytics to Improve Outcomes

Your electronic medical records (EMR) system stores vast amounts of information on all your patient encounters, but has it changed your productivity levels or outcomes for the better? Across the country, hospitals and medical providers are rushing to implement EMRs and meet meaningful-use requirements. However, the promises of improvement haven't yet been fully realized. There's a range of reasons, but the answer to this dilemma comes in the guise of health care data analytics.

We need tools that will aggregate the available information and develop predictive models for bedside use. Health care data analytics refers to software tools that analyze these troves of information -- often called big data -- into usable material. This process assists with comparative-effectiveness research and allows stakeholders to share results much faster and at a lower cost than traditional research studies.

Why Do We Need to Share Information?

The key to making sure that data captured by EMRs and other health care databases meaningfully affects patient care is simple: share it. Luckily, the Affordable Care Act's requirements are already pushing the industry in this direction. Sharing needs to happen not only at the local level, but also at the national level. Primary-care physicians share patient information with specialists, and hospitals pass on the needed information to primary-care providers. This communication gives each provider a full history of the patient, along with providing a more seamless experience.

On a larger scale, sharing data is necessary for the future of research. Clinical trials have limitations; take cancer trials as an example. Less than 3 percent of the patient population is represented in those trials, as the Journal of Oncology Practice reports. By using real data on patient experiences, we can get more detailed, more personalized views of which treatments work and find gaps to explore in future studies.

Current Limitations

Although most providers realize the benefit of sharing information, we still face many limitations -- particularly a lack of EMR standardization across locations and information-blocking practices.

Traditionally, EMRs have been developed independently by different vendors, and they don't talk to each other easily. In addition, different locations use different terminology and vocabulary for health-outcome data and inputs. For example, one hospital may enter date of birth as "Birth_DT," while another location uses "Date_of_Birth" and a different database uses "DOB." This lack of standardized language poses real challenges to data aggregation.

States are also making efforts on their own to share information. All states have some type of health information exchange (HIE). For example, Dignity Health participates in a private HIE, as well as a national HIE. With different states and different hospitals using different HIEs and EMRs, there's clearly a need to bridge the gap in some respect. As a potential solution to this issue, the Office of the National Coordinator for Health Information Technology (ONC) is supporting providers through its 10-year interoperability road map. The goals, which the organization aims to complete by 2024, are to increase transparency, assist with payment reform, and accelerate research. As part of its interoperability efforts, the office is also working on developing core terminology and vocabulary to provide a set of guidelines for the industry.

Even with better standardization, providers face the issue of information blocking. This refers to "an obstacle to the electronic sharing of a person's health information," as U.S. Senate Committee on Health Chairman Lamar Alexander put it at a hearing in July 2015. This blocking can be intentional or accidental and includes practices such as vendors creating barriers to sharing information with competitors, hospitals charging large fees to share records, or physicians outright refusing to send information for privacy reasons. Congress has been working on efforts to reduce this type of action, especially from vendors.

The Future Direction of Health Care Analytics

Many organizations, including Dignity Health, have developed their own HIEs and are actively working to share information that will support physicians in care delivery. Meanwhile, the National Patient-Centered Clinical Research Network (PCORnet) is making strides in comparative-effectiveness research by analyzing and interpreting big data. The network pulls data from Clinical Data Research Networks (CDRNs) and Patient-Powered Research Networks (PPRNs).

CDRNs compile valuable information from routine patient visits, while PPRNs pull together patient-generated data such as personal health records. The network also uses insurance-claims data and other information with the hopes of developing predictive models and clinical-decision support tools. PCORnet is still in the early stages, with hopes to expand in 2016.

We now have a range of tools to gather valuable information about patients that can quickly and affordably create major advances in patient care. There is still work to be done on standardization and intentional information blocking for competitive reasons. With advances in health care data analytics and more openness throughout the industry, big data has the potential to deliver on some of the early EMR promises of productivity and improved outcomes.