ElderBranch interviewed Dr. Caterino to discuss his paper, “Discordance between patient report and chart review of risk factors for antimicrobial resistance in ED patients.” Dr. Caterino wrote the paper, along with Lauren Graham of the Ohio State University College of Medicine.
What led you to research this topic?
This came about because emergency medicine physicians are under significant pressure as far as the volume of patients seen and the prolonged wait times in emergency departments (EDs). We have to work efficiently at seeing patients, evaluating patients, and appropriately diagnosing and treating them.
Older adults are particularly challenging because of their complexity. They have multiple medical conditions; many of them have been in the hospital or the nursing home recently. Even with electronic medical records, they have very extensive medical histories. And the emergency department physician is often seeing them for the first and only time so there is no history between the physician and the patient.
We obtain information by a combination of ways – one way is through chart review (looking through their medical records). As you can imagine with the amount of patients seen in the emergency department sometimes we can’t go through all of the medical records for each patient during their visit. At other times we obtain information through patient interview. When we talk with the patient in the ED, we try to get important facts out.
What we wanted to do in this study was determine where the information that we need to make important decisions resides. No matter how good your electronic medical record is, it is often incomplete. And often times the patients report things to us that are not in the medical records. We wanted to quantify that and compare it and see if we get more information from the [patients] or the electronic charts.
We focused on risk factors for antibiotic-resistance (AR) because of increasing problems of antibiotic resistance in our patient populations and the implications that has when we choose antibiotics. If we know someone has risk factors when we are choosing antibiotics, we want to take that into account when we make prescription decisions.
Please describe your study.
We enrolled emergency department patients and asked them if they had any of the risk factors (things like recent antibiotic use, recent hospitalizations, and certain diseases that make you more likely to have resistant bacteria). After the patient visit, we went back to their chart to see how many of these factors were documented in the chart prior to the visit.
What were the key findings from your research?
Interestingly, what we found was that patients were able to report the presence of risk factors more commonly than we found in the chart.
That probably indicates a couple things:
- In the ED, patients we see don’t always go to the same hospitals. So in some cases, care they received at other facilities doesn’t make it into the patient record so we need to rely on patient interview. That is particularly important in the ED.
- There were also specific factors – things like recent antibiotic use and recent surgery – that were not in the electronic medical record that we found during the interview.
In 10% of patients, they reported a risk factor that was not found in the chart. So if those people needed antibiotics, it may have changed our antibiotic selection. On the other hand, a lot of their medical conditions are well documented. If they have diabetes, or cancer, or immune system suppression that is usually in the chart and we can rely on the chart for that.
What are the implications and recommendations as a result of your work?
This has ramifications for our clinical practice. There are certain pieces of information where we still need to ask the patient and cannot rely on the chart. The electronic medical record is not sufficient for all conditions yet and shouldn’t be taken as the last word.
Secondly, from the standpoint of our research studies, when we are looking back at the charts, we may be missing information. So when we are designing research studies we need to be aware that there is additional information [that may be pertinent] that is not in the charts.
Are we ever going to get to a point where there is a universal electronic medical record that travels with the patient so when they show up in the ED you can see their complete care history no matter where they have been?
There is a long way to go, but what you ask about would be ideal. There are a lot of barriers. There are different systems that don’t talk to one another. There are some places that don’t have complete EMRs and they are very costly to implement – places like nursing homes. And there are concerns over privacy and exchange of information. I think the unintended consequence of the HIPPA Act is that people are afraid to share information even when it is appropriate.
In some cases, we are starting to get to the point where we have agreements with other healthcare systems that use the exact same record system. But that still doesn’t encompass urgent care, nursing homes, and institutions that are using other EMR systems.
You mentioned that nursing home residence was a risk factor domain for your study. Can you describe how living in a nursing home increases AR risk?
Many studies have shown that nursing home patients are more likely to have drug resistant organisms; there is a whole variety of reasons. They tend to be a sicker population with more medical problems, they tend to get more antibiotics than someone living in the community, and they tend to have more medical devices such as urinary catheters or central lines. They also have more medical procedures because they are sicker. All those sort of things add up.
Just being in the nursing home or even in the hospital, there is a chance of spread of drug-resistant organisms from patient-to-patient. I think we are doing better than we were previously, but that threat is still there.
In nursing home patients in particular, it’s important that we have as much information as possible about their risk factors for their drug resistant organisms so we can choose the appropriate antibiotics when we see them.
What advice would you have for nursing home patients or the nursing homes themselves on how they can a) reduce their AR risk, and b) make sure their physicians are aware of their AR risk?
There is good evidence out there that communication is poor between nursing homes and emergency departments in both directions.
What we need are defined transfer forms that have all of the relevant information including things like the accurate information on the patient’s desires for care (code status), medical conditions, perhaps risk factors for antibiotic resistance organisms, recent medication lists, their baseline status, including their baseline mental status, and what has changed that caused them to come to the emergency department.
There are some good studies out there on what those transfers should look like, and I think the one thing my study shows is that we can’t rely on our hospital’s EMR system in the emergency department to provide all of the necessary information because it is going to be incomplete.
It reinforces the need for very good, very informative transfer processes from the nursing home to the emergency department. And I think it is also true the other way. We in the emergency department don’t always do a good job of saying what happened in the emergency department when patients leave for a nursing home or another area of the hospital.