Nurses vs. Computers: Predicting Risk of Patient Harm

Not All Signs of Potential Harm Are Quantifiable

From chego101, via Flickr

Hospital nurses have many, many responsibilities and tasks, but one of the most important is to ensure patient safety by assessing patients for changes that can signal worsening of a condition or a new potential harm. Creating special units like ICUs, recovery rooms, and step-down units; flags on charts; various alarmed monitors; and safety huddles are a few of the ways hospitals have tried to identify potential problems. Now we have computerized tools to do this—or do we?

The complaint I have heard most from nurses about the electronic health record (EHR) is its inability to capture all the nuances of patient care or various patient problems, especially those that don’t involve easily quantifiable measures like heart rate or lab values. (For more detail, read our November 2016 report on nurses’ concerns with EHRs.)

One cannot accurately use a check mark to convey certain patient behavioral parameters or the “can’t put my finger on it but something’s going on with this patient” assessment that experienced nurses often make. In the April issue of AJN, we published an important study that investigates just this issue: “Identifying Hospitalized Patients at Risk for Harm: A Comparison of Nurse Perceptions vs. Electronic Risk Assessment Tool Scores

Deciding Whether to Implement an Electronic Risk Tool

Researchers at a large hospital system that was contemplating implementing an electronic risk tool into the electronic health record wanted to evaluate how the tool compared with nurses’ judgment in identifying patients at risk for harm. They compared nurses’ judgments of risk (which came out of twice-daily safety huddles) with the identification of risks by the risk assessment tool. The data were collected over a three-week period on three different nursing units.

What they found—and I bet most nurses will not be surprised by this—is that nurses perceived risks differently, often noting risks where the tool did not. From the abstract:

 “In 746 data pairs, differences between the nurses’ harm risk perceptions and the electronic tool’s harm risk reports were statistically significant, supporting our prediction that there would be no correlation. The most significant difference was seen in instances when a nurse identified a patient as being at higher risk than the electronic tool did, often citing behavioral or psychosocial issues as the reason for concern.”

This wasn’t just hypervigilance on the nurses’ part. They were often correct in their assessments, finding real issues that were not detected by the tool. Because of the results of this research, this hospital system chose to not implement the tool.

We hope you’ll read the study (CE credits available); you can also listen to a podcast discussion with the authors (available, along with many other podcasts, at our Behind the Article podcasts page).


Editor-in-chief, AJN

One Comment

  1. Eunbi Anna Ko July 14, 2017 at 9:32 pm

    Hi, as a student nurse currently learning the in’s and out’s of working on the floors with the use of modern technology and good old “nursing tricks”, I found your article quite interesting. I wondered how efficient the electronic health records were with pre-filled and pre-designed tools such as risk assessment tools. Because us humans can make mistakes, I think it’s nice to have a guide to follow as a reference but I believe it’s also important to utilize our nursing knowledge based on experiences working as a competent nurse. I believe the implementation of utilizing an electronic risk tool should be based on evidence based research with a combination of incorporating the nurse’s knowledge and judgment to perform such assessments for the best patient outcome and results.

Comments are moderated before approval, but always welcome.

%d bloggers like this: