Close Server: KOPWWW05 | Not logged in

Hot HIM Apps

Computer-Assisted Clinical Trials

Patient finding goes real time.

Computer-assisted coding (CAC) and computer-assisted CDI are becoming mainstream in healthcare. Natural language processing (NLP) technology is layered atop EHRs and digital documentation to flag specific words, terms and phrases triggering code assignment in CAC applications, and case identification in CDI. Now the same technology is being applied to clinical trials, research and advanced treatment protocols.

NLP electronically "reads" patient information to find in-house patients matching specific trial and treatment protocols. It screens patients and alerts researchers-allowing them to begin the patient interview process real time. Here's how it works.

Understanding the Workflow
Instead of sifting through monthly IT reports and manually reviewing records post-discharge to find candidates, researchers are "alerted" upon admission while patients are still in-house and in a hospital bed. There are four steps to the workflow. NLP electronically reads all clinical data and documentation upon patient admission. The system electronically screens potential candidates for clinical trials, research or advanced treatments using "rules" already built into the system by nurses and researchers. Armed with a targeted list of potential candidates, nurses interview patients, check criteria, and initiate trial participation discussions with clinicians and patients. Trials, research and advanced treatments begin much quicker for patients meeting the research criteria-saving time, dollars and perhaps lives. One organization, Austin Cancer Centers in Austin, Texas, is using NLP to determine which prostate cancer patients are best suited for advanced treatment protocols within their oncology service lines.  

Advancing Cancer Care
What formerly took Austin Cancer Center's research group an entire day and multiple team members to accomplish is now done within a few minutes every day. The team is able to screen more qualified prostate cancer patients using fewer people-and achieve better clinical outcomes. For example, if an organization sees 500 prostate cancer patients per week and reviews documentation weekly to screen for candidates, perhaps 50 will meet the basic inclusion criteria: which means the manual review of many records. However, if charts are screened daily using technology with inclusion and exclusion criteria, perhaps only 3-4 candidates' charts need further review. It's far more effective to review 3-4 per day, than make time for 50 reviews once per week. The final step is to qualify or disqualify the patients for the trial, research or treatment. Once done, the organization has identified and signed up the exact 10 prostate cancer patients best suited for the new drug, research or treatment with fewer staff, fewer steps and better physician satisfaction. All steps are done while the patient is still in-house and easily accessible.

Computer-Assisted Clinical TrialsOvercoming Documentation Hurdles
Because much of the important clinical documentation for screening patients is still paper-based, it is often overlooked during the process. Poor physician documentation sometimes erroneously disqualifies patients from a trial, research or advanced treatment. By using NLP to read documentation, screen candidates and alert researchers, more patients have the opportunity to participate-and possibly be cured. The next step for computer-assisted research lies in the ability to identify, screen and enlist patients across an entire health system or region. Stay tuned for Austin Cancer Center's next steps toward this goal.

Steve Bonney is executive vice president, RecordsOne.

Hot HIM Apps Archives

Completely agree on the use-case; but wondering, can Austin Cancer Center considerably improve on its NLP accuracy? The market is learning that not all NLP engines are alike. Opportunities for deeper pools, better pre-qualification, still fewer manual interventions.

Mark February 10, 2015


Email: *

Email, first name, comment and security code are required fields; all other fields are optional. With the exception of email, any information you provide will be displayed with your comment.

First * Last
Title Field Facility
City State

Comments: *
To prevent comment spam, please type the code you see below into the code field before submitting your comment. If you cannot read the numbers in the below image, reload the page to generate a new one.

Enter the security code below: *

Fields marked with an * are required.

View New Jobs, Events and More


Back to Top

© 2017 ADVANCE Healthcare, an Elite CE company