The National Center for Human Factors has influenced the front line providers within MedStar Health by encouraging them to think about errors. By looking closely at the ways the system might have facilitated an error and created a hazard,  we can then better design the system for future patients.

The case that has gotten the most attention since we started the Center in 2010 has been where a nurse misread a glucometer. As a result, she misinterpreted the patient's results of having high blood glucose when the patient really had low blood glucose.

The patient had to go to the Intensive Care Unit. Fortunately there was no adverse outcome for the patient. The patient was given sugar and did well afterwards. However, immediately after the incident, the review of the case made it look like the nurse had made a mistake. When there was a second, similar event that occurred, our team went and looked it an determined that the real root cause of this error was the design of the glucometer.

In 2012 a highly-motivated nursing director at MedStar Washington Hospital Center wanted to improve patient experience. She came up with this idea of having the nurse managers round each day. It's advised by a bunch of outside boards as a best practice.

She wrote a bunch of questions on a sheet of paper and the nurses would ask these questions each day of each patient inside the unit. The stack just kept growing.

She came to MI2. We not only made her form electronic so she could roll it up and see the data, but we helped employees on the front lines by providing them information and little fixes that could help make the experience even better for them and better for the patient.

There's three different ways we were able to do this. One of the ways we're able to do this is take information out of the electronic medical system and provide it to that person right at the point that they're giving the survey to help the patient understand if they have a deficiency in their knowledge.

For example, one of the questions that is high on patient experience scores is, “Do you know the medications you're taking and the potential side effects?” We put that question there, but also directly under it, we listed the medications that the patient is taking. There's no clicks, there's no nothing. It's right there.

If the patient says, "I don't really even know what medication I'm taking," they nurse can say, "You're on Rocephin. That's an antibiotic. You're on Vancomycin, that's an antibiotic. It can cause flushing." We're able to embed the correct information right at the right spot so that our associates actually look a lot smarter and the care looks like it's very coordinated and everyone knows what everyone else is doing.

The second thing we were able to do is fix problems in real time. We created hashtags within the forms, so if the person who is administering the survey finds a problem they can mark it up with a hashtag so that the system can do something about it.

The nurse administering the survey to the patient can write "the toilet is broken," add a #facility after it and when she submits it, the facility person is notified.

The third way we added value is to aggregate everything in a real time dashboard. We’re able to look across an individual unit, an institution across the whole enterprise, by day or by hour. We could actually look in real time at what the patient experience scores are doing at a particular unit.

There's several ways that we really bridge the gap between different kinds of industries and one of those is the area of visualization. In healthcare there's a lot of data. It could be patient data, it could be patient safety data. It could be the different shifts of different kinds of employees. Integrating that data to make sense of it and to really understand where safety hazards reside, is something commonly done in aviation and commonly done in defense, but not so common in healthcare.

One of the big contributions our group has made is developing some really dynamic visualizations that allow people to see these patterns more easily. We're also working towards building automated methods for extracting trends from those data.

The big goal there is to make it so that any safety hazards that exist in the MedStar system, and really in any healthcare system, will float to the top and become much more transparent so that we can act and make the system safer.

In the early 2000s, we developed a gate device that is ceiling mounted and has a motorized component to it that enables people to be in a harness working with a therapist without the fear of falling. This is great or patients that are newly injured, and complete spinal cord patients, any diagnosis where you may be relearning after an injury.

The gate became a commercially available product and we gave the rights to the engineer because we didn't have the infrastructure to take it to market. Now our product is in a third of the VA Hospitals. It's in a number of rehab hospitals. We're very proud of our participation and our work with that tool and I think it's great that it's widely distributed.

We didn't have the infrastructure at that time to look at it from a business perspective and today we would. We have lots of new ideas for treatment techniques and equipment, and when we find promise in those we're working through MI2 to make sure that we're protecting our intellectual investment.