AP and CP reporting, from interfaces to IT wishes

Create PDF

March 2024—Anatomic and clinical pathology reporting—what’s working, what’s missing. Three pathologists (all board certified in informatics) and representatives of three information system companies met online Dec. 19 with CAP TODAY publisher Bob McGonnagle to talk about reporting needs and what’s optimal. The first half of their discussion was published in the February issue, with CAP TODAY’s guide to anatomic pathology computer systems. The second half begins here.

Joe Nollar, can you comment on customized reports?
Joe Nollar, associate vice president of product development, XiFin: We need to help pathologists create reporting products that assist treating physicians in understanding the diagnostics better, which will lead to better patient care.

Another trend I like is patient education materials at the end of a report that help explain the test results and what they mean to the patient.

Sharing surgical pathology reports with patients is now a requirement. Diana Richard, what are your thoughts on this?
Diana Richard, senior director, pathology and strategic development, XiFin: There’s a value provided to ordering physicians, specialty-specific providers such as oncologists, and patients. While patients may not understand the full context of the reports they receive, transparency will drive inquiry and presents an opportunity for pathologists to reinforce themselves as contributors to the diagnostic and treatment process.

Richard

As we talk about combining routine AP and clinical testing and result reporting with new, complex molecular testing, there is also a value to the payer. These tests are expensive and we can’t afford to perform them for free. Payers will often deny these claims and request additional information on new testing methodologies to determine if medical necessity for those services is justified. The more comprehensive the report is in reconciling the utility and value of what was performed, the more successful we are in collecting on the services rendered. Long term, these success rates play a substantive role in driving payer policy. Comprehensive reporting creates a horizontal value proposition to the provider, patient, payer, and pathologist.

Ross Simpson, there’s an idea here that a pathology report could justify reflex testing or be used as a justification for approvals, whether for drugs or further testing. What do you think about those ideas?
Ross Simpson, MD, head of pathology informatics, Methodist Hospital, HealthPartners, St. Louis Park, Minn.: I think it’s great. We do mostly manual reflexes, but we are able to reflex tests from the AP section. We can order a test and it automatically orders a CP test. We moved our FISH evaluation to a CP test because it makes sense for us. We put that back in the AP report as a link, so that’s where you get AP and CP coming together.

I worry about custom reports. Think about an airplane cockpit—would you let every pilot have a custom cockpit? No. I like the idea of having a report that’s well made and agreed on. Specialties need a different look at reports, but internally we dissuade people in our group from doing anything custom because we change lab tests. The maintenance when you get down to individuals is too high. I’m for unified reports that are well designed, and that’s committee work and a work in progress.

Dr. Simpson

Sometimes when new groups talk about going to discrete reporting, they’re doing two things—a narrative report and the discrete results, and that’s another mistake. You don’t see that in the clinical lab. You don’t write the sodium is 146 and then put in the value of 146. You just put in the value of 146. If you’re doing discrete reporting in anatomic pathology, that should be it. You put out the synoptic and you don’t need a narrative.

We have done work where we take synoptic data and create a narrative. At least then it’s in sync with the synoptic, and the pathologist doesn’t have to do it. But you want an efficient and accurate input and an efficient, accurate output. Discrete data is perfect for that. We have a value; we know exactly what it is and we can link to outside websites or whatever we need, consistently.

When I think about the physicians treating the patients on whom you’re sending reports, I envision a Gaussian curve on which there’s a percentage who know how to make sense of the report, some who more or less know what’s in the report, and some who are falling behind, who just want to know, What should I do now? I’m sure you get those phone calls as a pathologist, is that right?
Dr. Simpson (HealthPartners): We do. We handle some of that in comments—we attach our comment to the diagnosis field and don’t allow it to be separate. It has to be sent with the diagnosis everywhere it goes. If we have a clinician we think is external to the process normally and might not know what to do, we will provide information on what the next step would be or what we recommend.

I’m in an integrated group, so it’s 500, 600 doctors, and everyone knows where to send someone. When we know it is a malignancy, for instance, we’ll copy the oncologist who is going to get the call on that report even though it came from a family practice physician. We know the flow.

Are clinicians benefiting from the structured and discrete reporting?
Dr. Simpson (HealthPartners): Yes, they’re excited about it. The breast working group is excited about the data they’re getting from the discrete report.

Alexis Carter, do you still have concerns after hearing this discussion?
Alexis Carter, MD, physician informaticist, pathology and laboratory medicine, Children’s Healthcare of Atlanta: I’m going to have concerns until I see the end product. But I’m a molecular pathologist. The way I’m seeing people try to split apart molecular data—we need to do a lot of work to make sure that that data will display appropriately, especially if they try to put them in integrated reports.

What kind of a comment around consultation or case comments are you using as a practical matter to get through your casework? How do you display it on the report and what are the supplementary steps you take? Follow-ups with phone or email, or fielding questions?
Dr. Carter (Children’s Healthcare of Atlanta): We have a tight-knit group of physicians. If, for example, my molecular does not match the diagnosis, then I call the ordering physician before I issue a report. A lot of what I do happens before I issue the report. Now, with patients getting our reports almost as soon as we sign them out, we try to coordinate before signing out reports so there is a plan for the patient before they get it. For example, if I run into a case where there’s a 90-plus percent chance a child has a germline variant for a hereditary cancer predisposition syndrome, I call the provider before I sign out the report and let them know they need to get germline genetic testing because there’s a good chance it will be positive, and this will impact the patient’s long-term care.

Dr. Carter

I agree with Ross—I do not like the idea of ordering physicians having customized reports. When you go to a store’s website to find where its brick-and-mortar locations are, most often the store locator is at the top right of the webpage and has a standard location symbol. This standard design is done for usability reasons. Pathology reports are no different. They need to be constructed in a standard way across practices and systems so people know where to find the information they’re looking for. For example, the patient information, case number, et cetera, are at the top; the next is the “Final diagnosis” or “Interpretation,” which is usually in a big bold font so the eye is drawn to it. I feel the same about molecular data. Standard displays make sure people are able to see the important stuff quickly. When health information exchanges or integrated reports disaggregate our data and display it in a way that no one will be able to make sense of, or worse, they will interpret it incorrectly, they’re doing the wrong thing for the patient and making things less safe for them.