Readers Write: Natural Language Processing: Putting Big Data to Work to Drive Efficiencies and Improve Patient Outcomes
Natural language processing (NLP) is increasingly discussed in healthcare, but often in reference to different technologies such as speech recognition, computer-assisted coding (CAC), and analytics. NLP is an enabling technology that allows computers to derive meaning from human, or natural language input.
For example, a physician’s note may state that a patient “has poorly controlled diabetes complicated by peripheral neuropathy.” When notes are analyzed through an NLP system, coded features are returned that can:
- Suggest codes such as ICD-9 or ICD-10 that may feed a CAC billing application;
- Classify a patient according to applicable quality measures such as poorly controlled diabetes mellitus, to support a reporting tool;
- Populate a data warehouse;
- Feed analytics applications to support descriptive or predictive modeling, such as the likelihood of a patient being readmitted to a hospital within 30 days of discharge.
Healthcare is data intensive from both clinical and business perspectives. While the industry’s transition to electronic data collection and storage in recent years has increased significantly, this has not actually forced physicians to code the majority of meaningful content. Eighty percent of meaningful clinical data remains within the unstructured text, as it does in most industries. This means that it remains in a format that cannot be easily searched or accessed electronically.
NLP can be leveraged to drive improvements in financial, clinical, and operational aspects of healthcare workflow:
For financial processes, automating data extraction for claims, financial auditing, and revenue cycle analytics can impact the top line. NLP can automatically extract underlying data, making claims more efficient and offering the potential for revenue analytics.
For clinical processes, automatically extracting key quality measures can support downstream systems for reporting and analytics. NLP can infer whether a patient meets a quality measure rather than requiring individuals to manually document each measure for each patient.
For operational processes, descriptive and predictive modeling can support more effective and efficient operations. NLP can extract hundreds of data elements per patient rather than the 2-4 codes listed in claims, producing better models and supporting business insight and diversion of resources to high risk patients.
So, NLP is a powerful enabling technology, but it is not an end user application. It is not speech recognition or revenue cycle management or analytics. It can, however, enable all of these.
There is a battle underway that is increasingly recognized in the healthcare space. Individual hospital divisions seek turnkey solutions and frequently purchase NLP-enabled products. But at a broader level, health systems as a whole do not want to pay repeatedly for similar technology. They seek best-of-breed infrastructure, wanting a combination of electronic health records, data warehouses, NLP, and analytics.
This battle will increasingly highlight best-of-breed data warehouses, data integration vendors, and natural language processing technologies as health systems search for a scalable, affordable, and flexible healthcare infrastructure to feed a suite of clinical, operational, and financial applications.
Dan Riskin, MD is CEO of Health Fidelity of Palo Alto, CA.