FacebookTwitterLinkedIn
  • Blog
April 16, 2018

Three Practical Applications for Natural Language Generation in Healthcare

By: Mellisa Udhayananondh

With 96 percent of all hospitals in the U.S. having a certified electronic health record (EHR) in 2017, clinicians are operating in an incredibly data-rich environment.  Rarely within the walls of today’s modern hospitals does an action go undocumented.

This data is often stored in coded and structured forms (in the patient record, drug databases, medical terminology databases), best used for slicing and dicing with precise queries and quantitative analysis. Yet, these sorts of analyses – while often crucial for patient care – may not be part of the skill set of physicians or nurses, whose primary responsibility is caring for the patient. How then can hospitals make this wealth of data and information more accessible and actionable for clinicians to improve patient outcomes?

One avenue is through natural language generation – turning this trove of data into stories, a medium easy to digest and familiar to clinicians whose preferred method of documentation is language. For clinicians who lack the time to perform deep data analyses, stories offer a way to learn insights about their clinical data, allowing them to improve the quality of care for their patients and reduce overall costs for a hospital.

Some practical applications of natural language generation for clinicians include:

1.  Improving clinician KPIs

Clinicians, exhausted by long shifts and managing multiple complex patients, don’t have time to interpret dashboards.  Even though they can detail important care measures such as average length of stay, rate of medication reconciliation compliance, and barcode-scanning compliance.

By adding plain-English stories to these dashboards that highlight only the most important information – which KPIs to focus on and what next steps a clinician should consider in order to improve a KPI – the information becomes much easier to comprehend than looking at graphs or charts alone. Additionally, improving these KPIs is important not only to improve the quality of care for the patient, but to comply with national standards for government reimbursement.

2.  Clarifying patient bills

As hospitals vie for patients, natural language generation gives organizations a competitive edge by creating a bill that’s transparent and easier to read, thus enhancing the patient experience and encouraging patient loyalty. By turning billing codes into natural language, with explanations about charges and the relationship to diagnosis, patients will feel better about paying their medical bill – even as it arrives months after a hospital encounter.

3. Personalizing research study recruitment

With the wealth of information collected on each patient, hospitals now have an easier time finding individuals for various studies. Actually getting them to become participants? A different story. Personalizing this outreach effort through stories tailored for each potential participant can help enrollment and recruiting efforts.

Good communication is vital in the healthcare sector, and narratives can help explain the information in structured databases more clearly than information presented in charts, graphs, or even flowsheets. For hospitals considering ways to use this information to power their AI strategy, natural language generation is an essential technology.

Conclusion

Regardless of how hospitals and health system are capturing data, it is an industry ripe for disruption when it comes to actually understanding it.  Click below to learn more about Quill, our intelligent automation platform that can put natural language in reports and dashboards.