A feeding tube that's malpositioned, a small but new pneumothorax, a newly discovered pulmonary nodule -- these are the kinds of radiology-related critical test results that should be communicated in a timely and verifiable way to the doctor caring for the patient. But they can often be a bane to radiologists.
If only we had the equivalent of the Batphone that Commissioner Gordon could use anytime to reach Batman directly -- but in our case it would be to reach the patient's doctor. Most radiologists are not hesitant to speak with their referring physicians. On the contrary, we value being able to contribute directly to patient care by discussing results with our referrers in direct conversations and in conference settings.
Rather, there are two main barriers to communication that too often stand in the way of optimal -- and sometimes even required -- communication. These two barriers can be summed up as "Whom do I call?" and "How do I reach her or him?"
Who should get the results?
At first glance, the first question may seem simple to answer. As a general rule, the doctor who ordered the test is responsible for receiving the results. However, in the outpatient setting, that doctor may very well be on vacation, and it may take some painstaking sleuthing to determine who is covering for him or her. In the inpatient and emergency settings, doctors work shifts of eight to 12 hours. It's very common for the doctor who ordered the exam to be off-shift by the time it's performed and the radiologist starts interpreting it.
Dr. David Hirschorn from Staten Island University Hospital.Thus, in these settings, more often than not the ordering doctor is not the one who needs to be alerted to urgent or unexpected critical findings. This is the first cause for the radiologist's hesitation in making the phone call: She or he does not even know whom to call. In most U.S. hospitals there is no quick and easy way to determine which doctor is currently responsible for the care of the patient.
It often involves a lengthy attempt to reach a nurse on the ward who will look the number up on a grease board and may or may not respond with the correct information. The frustration only builds when the radiologist tries calling that phone or paging that pager only to get no response, or for that doctor to reply, "That nurse was mistaken; call her back and ask again because that's not my patient."
Even when it is clear who needs to be contacted, the next battle is to determine how to contact the doctor. The phone number found on a list or given by the operator might be out of date, or that phone may be out of order and the doctor is using a different one. Moreover, some medical services have different numbers to call depending on the time of day or day of the week.
Faced with these hurdles, it's tempting to try to rationalize that perhaps special communication is not really necessary. This decision of course is not in the best interest of patient care, and in some circumstances, it can lead to medicolegal exposure for failure to communicate properly, should it lead to an adverse patient event. Guidance on which findings need to be communicated in a manner beyond the written report can be found in the American College of Radiology (ACR) Practice Parameter for Communication of Diagnostic Imaging Findings.
Alternative approaches
Several solutions have been developed to deal with this challenge. Most of them start by allowing the radiologist to click a button of some sort to electronically flag the radiology report as requiring special communication. One such system sends an alert to the referring physician's smartphone, or all those on the physician's team, specifying the critical finding or the entire impression of the report.
It requires the recipient to acknowledge receipt, and it will escalate the handling of the message over time until the message is acknowledged. This may work well in settings where most physicians have a smartphone and are willing to install the necessary app.
Another system sends all of the flagged reports to a clerk who then takes the time to locate the appropriate referring physician and read the report impression to the doctor over the phone. At my own hospital, we have the clerk tackle that difficult task of locating the right referrer, but the clerk then connects the referring physician to the radiologist so they can have a two-way conversation. When the report is flagged, the radiologist indicates if the finding is urgent, or whether it's just important but not urgent.
Urgent findings are those that require the doctor to be reached within an hour, no matter the time of day. On the other hand, important findings such as a pulmonary nodule can wait up to two days and are pursued only during regular business hours.
Nothing slips through the cracks
The key to success of this and, indeed, of all of these methods is that there's a database that tracks all of the messages and ensures that none slip through the cracks. Before the launch of this system, our radiologists knew that they could ask the clerk to try to get a doctor on the phone for them. However, should that clerk forget, or even try very hard but not succeed and then go off shift, the request would be lost to follow up.
As a result, radiologists were hesitant to ask for the clerk's help. But now that they are certain the request will not get lost, and they can actually see the list of all of their open and recently closed requests and the full audit trail of how the request was handled, they no longer hesitate to ask for it. In fact, they don't even need to pick up the phone; the request has been reduced to a button-click on our dictation system.
From the radiologist's point of view, she or he clicks the request button, specifies if it is urgent versus important, clicks OK, and then can continue working on the next case. Typically about 10 minutes later the phone rings: It is the clerk saying she has the doctor currently caring for the patient on the line, waiting to speak with the radiologist. This is an example of one method that has drastically reduced the barriers to communication and fostered many more two-way conversations between referrers and radiologists in which results are not just delivered but can also be discussed.
Strong critical test results management systems smash communication barriers, increase radiologist participation in patient care, ensure better patient outcomes, prevent medical errors, and decrease medicolegal risk. A variety of these systems exist today, some commercially available and some developed in-house at academic institutions but which can be replicated. Every radiology practice should evaluate which method best suits its needs, but the time has come to move forward with some solution to keep track of all of those critical results and ensure their delivery.
Interested in learning more? I will be delivering a full lecture on this topic, replete with screen shots and real-world statistics, at the New York Medical Imaging Informatics Symposium (NYMIIS) on September 21, 2015, at the New York Marriott Marquis. Details are available at www.nymiis.com.
Dr. Hirschorn is the director of radiology informatics at Staten Island University Hospital.
