Published in the May 2013 issue of Today’s Hospitalist
HOSPITALISTS ARE IN THE RIGHT PLACE at the right time to do something not normally associated with inpatient care: diabetes screening and prevention.
A study recently published in the Journal of Hospital Medicine showed that when hospitalists take advantage of “opportunistic screening,” they can achieve noteworthy results. In the study, 24% of 275 patients admitted for elective orthopedic procedures who were screened when they came in NPO were found to have previously unrecognized inpatient diabetes or impaired fasting glucose. Even more startling, all of those studied were insured “and more than 95% of them had seen a primary care provider within the past 12 months.
Those results raise questions about the effectiveness of outpatient diabetes screening. But they also suggest that hospitalists may have an inside track to reach the 40% of patients with diabetes who go undiagnosed, says lead author Ann M. Sheehy, MD, MS.
“We have to get creative about meeting patients where they are.”
Filling this new role wouldn’t be costly or time-consuming and could help prevent years of high health care costs due to complications from elevated blood glucose, Dr. Sheehy says. She is clinical associate professor and head of the hospital medicine division at the University of Wisconsin School of Medicine and Public Health in Madison, Wis.
Screening patients who come in for elective surgery also may be an important move within accountable care organizations (ACOs), which give all providers “skin in the game” when it comes to prevention.
“Many patients do not seek regular care from their PCP in the ambulatory setting,” she says, “but medical homes and ACOs have an opportunity to facilitate preventive care or screening, whether that’s done as an inpatient or outpatient. We have to capture the opportunity and get creative about meeting patients where they are.”
Dr. Sheehy talked to Today’s Hospitalist about that opportunity and its potential impact on diabetes diagnoses.
Why did you choose to study this group of patients?
Hospitalists collaborate a lot with orthopedics. For elective orthopedic patients, it’s easy to know if patients are coming in for surgery on a certain day NPO and when they come in for follow-up. Everyone in this group gets a chemistry panel, which includes glucose. That gives us multiple looks to see if a patient’s glucose is elevated.
How did you ensure that elevated glucose really was diabetes? We were always taught that in “hospital blood sugar tests-even a glucose of greater than 200 “were not reliable because patients are subject to stress or we can’t be certain when they last ate. For this study, we were able to capture a fasting blood glucose because elective surgery patients came in NPO. However, we did not know if this value might be affected by stress.
So we wanted to re-test six weeks later, when patients were more comfortable, to validate the preoperative glucose as truly unstressed. That also complies with the American Diabetes Association’s (ADA) mandate of having two values to diagnose diabetes.
Were you surprised by the degree to which diabetes was being missed in the outpatient setting?
Twenty-four percent was much higher than I thought it would be, given that all the patients had insurance and the majority had a recent primary care visit. Our results are tempered by the fact that the majority of these patients had prediabetes: The 24% included prediabetes and actual diabetes, but 88% were in the prediabetes range, 100 to 125 mg/dL.
Part of the reason may be that PCPs use screening criteria issued by the U.S. Preventive Services Task Force. In a previous study, we found that using ADA guidelines picks up many more patients with diabetes. But we also believe that provider and patient awareness of diabetes screening is lacking. Patients all know that they need their blood pressure and cholesterol checked at the doctor’s, but you don’t have the same awareness regarding diabetes.
So should just having a high preop number prompt a follow-up? We showed that a majority of patients “64% “still had elevated numbers postop, so the preop value is pretty reliable as a risk stratifier. You have to be cautious, but a positive result should prompt a follow-up glucose in a nonstressed setting.
Why are hospitalists the go-to providers for this?
It is similar to administering vaccines in the hospital, and it makes sense to use the opportunities you have for preventive care. You don’t know when you will have that opportunity again with a particular patient.
But is it practical? How much more can hospitalists take on in terms of duties?
Hospitalists are responsible for other core measures, and this could be protocolized in similar ways. If the preop blood sugar is elevated, just put a note in the discharge summary and send a note to the PCP to redraw the test when the patient is unstressed. Hopefully, that will become more automated in the future through an electronic medical record.
For now, you just have to be more aware when a chem panel was done fasting and make sure undiagnosed diabetes and prediabetes are on your radar. You’ll be able to diagnose diabetes and prediabetes in two-thirds of the patients with elevated preop levels, so it’s a good practice.
How do you make sure you don’t repeat a test or tread on the PCP’s turf?
We want to avoid stepping on toes as much as we can. It gets back to communication. Say to the PCP, “This is what I’m seeing. Do you want me to put in the order or will you?” In our institution, we’re focused on that communication piece to find common missing pieces. It’s much more difficult if patients come from outside our system.
Have you changed your own practice based on your findings?
I do think about this differently. I am much more aware of high blood sugars and more likely to check POC glucose for 24 to 48 hours when I notice someone’s level is high. For example, I had a patient who had preoperative glucose of 150. I ordered POC glucose checks; most were in the 130 to 160 range. I checked a hemoglobin A1C and it was 6.8%. I notified the patient-who was asymptomatic “and his provider that he should have follow-up for diabetes.
I feel like I diagnosed diabetes in a patient for whom that diagnosis may not have been made for years. For the patient, this was a wake-up call. He had gained 30 or so pounds over five years, and he planned to make changes in his life because of this.
What does this mean for those without a PCP or access to care?
I suspect those people would have much higher levels, and more in the diabetes range. A patient with regular care would likely have a better chance of being diagnosed earlier.
What happened to the patients you diagnosed?
I personally called all patients who had two elevated blood sugars and told them to follow up with their PCP. Most were surprised or concerned. And I sent all PCPs a letter. For patients with findings in the prediabetes range, it’s more about monitoring and trying to avoid progression to actual diabetes via lifestyle changes.
Paula S. Katz is a freelance health care writer based in Vernon Hills, Ill.