Implementing Standardized Radiology Reports

Structured reports in radiology increase clarity for both radiologists and clinicians. They allow for better communication of vital patient information that may otherwise be buried, and that improved communication can result in speedier diagnosis and treatment.

Yet, there are no standardized reporting systems used in radiology departments across the country, leaving individual departments to develop their own. Diagnostic Imaging spoke with Alexander J. Towbin, MD, the Neil D Johnson Chair of Radiology Informatics at Cincinnati Children’s Hospital in Ohio, who helped develop the standardized reporting system that is now used at his facility.

What encouraged you to develop and implement a standardized reporting system for your department?

We first started thinking about it when we installed PowerScribe 360 in 2009. At that point, I started learning more about structured reporting through PowerScribe’s capabilities. And that brought me to thinking about the potential benefits of structured reporting. I brought this information to my chair and we talked about it as a leadership group.

How did you decide on the formatting of the reports?

The literature and detail about structured reporting that’s already out there shows that most people only consider the bulleted format as a true structured report, so that was the basis from where I started. But then as I thought about it and tried construct the reports, I had to consider their usability and how a completed report would look, how it was going to print out, and how it would come across different information systems. That led to different approaches.

Did the types of reports your department issues make a difference in their design?

Yes, there are differences between the complex procedure and the simple reports. As we built them, we would check to see how they would look when completed. For example, with more complex reports, such as CT of the abdomen and pelvis, we would list the different organs and add a comment on each one or say “normal,” and that works well.

But when you start bringing that type of format to a simple procedure, like a chest X-ray, it becomes more challenging. You could do the same thing, say “lungs normal, airway normal, bones normal” and you could go all through it, but to us that seemed a bit excessive for something like a chest X-ray. For those simpler types of studies, it made more sense to have a similar structure but in a prose format instead of bulleted.

I also think that longer reports that are in a bullet format tend to be easier to read. Studies have shown that people prefer that type of report. It’s easier to pick out the important information and to retain it.

How did you convince radiologists who were reluctant to change to standardized reporting?

[[{"type":"media","view_mode":"media_crop","fid":"21777","attributes":{"alt":"","class":"media-image media-image-right","id":"media_crop_6279411675020","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"1537","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"height: 150px; width: 175px; float: right; margin: 5px;","title":"Alexander J. Towbin, MD","typeof":"foaf:Image"}}]]We did that in a few ways. We started off with sort of a straw man by putting in standard reports that weren’t too structured. They did have some level of structure, but they weren’t specifically department designed. So, when the radiologists opened up the chest X-ray standard report, they became used to the fact that the reports were there. But in a sense, they weren’t happy with the reports because they weren’t designed for our department nor were they designed to be a long-term solution. This got them interested in trying to make the reports better.

Another approach we used was to try to get everyone involved in the whole process. We did that by creating a structured reporting committee and anyone was welcome to volunteer to be part of it. We looked for representatives from every department division and we expected them to create the reports for their divisions.

The reports were created and then were sent through an approval process. A working subcommittee worked on standardized language, making sure we were using similar terms throughout the reports, that the formatting was similar, and that the little details were the same, down to how many spaces you had after a period and the location of certain elements. There were negotiations between the authors and the subcommittee as they worked this through.

Once the authors and the subcommittee agreed on the reports, they were sent out to the appropriate departments. For example, the head CT report went to the neuroradiologists and the chest X-ray report to all radiologists. We had a one-week comment period so people could tell us what they liked or hated about the reports. We didn’t necessarily agree to or implement every comment, but all comments were addressed.

Finally, once we implemented the reports, we made it harder for the radiologists not to use it. The reports were pre-populated with questions and it was actually more trouble to delete the information already there than to just use it. We did also ask for continual feedback so no report is ever finalized.

Another part of the program’s success is that we incentivized its use. There were individual and department goals and the departments had to get 90 or 95 percent to meet their goals. That puts a bit of peer pressure on the group.

How long did it take for the process of developing the reports and getting them into circulation?

It is still ongoing but we’re in a quiescent phase right now. It took about 12 to 15 months before most of our reports were complete. We are still building new reports. We can now create specialty-based orders, such as an MR enterography, and indication-specific orders. Every time we create one of these new orders, we build on something older, such as an MRI of the abdomen and pelvis, but then we tailor it.

Was there a steep learning curve to get used to the standardized reports?

There is definitely a learning curve to authoring the reports, but as for using it, the learning curve is pretty quick. Most people understand them right away.

The hardest part of using the reports is knowing the choices in the pick list. You could have multiple choice style options that have been built in ahead of time and knowing that those options are there and what they contain is part of that learning curve. Otherwise, I think it’s a pretty easy system.

So why hasn’t this been done before then?

The technology wasn’t there before, but there are some other reasons why. The biggest reason is the argument of autonomy versus standardization. Most of us went into medicine because we wanted to be autonomous, free thinkers. We think we’re smart and we do great things for our patients, and the question becomes how can we structure that in standardized reports?

The reality is that we are all very good individually, but the problem is that collectively, no one understands us. So if I say something one way and you say something slightly different, how does the person on the other end know that we both mean the same thing. Maybe in a small practice, it doesn’t matter, but as our practices get bigger, as with the 36 radiologists in our department, it’s hard to know what one person means versus the other.

The other challenge is getting a department to agree to move ahead with these reports. In our facility, we were in a sweet spot. Our department size was big enough where we needed it, but small enough that we could do it. It could be harder in a larger department. It also takes leadership to do take this on.

Was it costly to implement standardized reporting?

I’d say not at all. But, if time is money, then yes because of all the people who took part and spent time authoring the reports. Two people in that subcommittee spent a lot more time working on it. But aside from the time is money argument, not costly at all.

What would you like your colleagues to know about implementing standardized reporting?

First, structured reporting is possible at every department level and using standardized structured reporting has everyone speaking the same language. It also brings value to the department. We are able to communicate more effectively with our clinical colleagues ago help guide treatment by providing actionable reports. I think those things are really key.