To the editor:
In response to the article “Point-of-Care Glucose Testing is Unnecessary in Non-diabetics with Normal Mental Status,” let me start by saying I understand the emphasis of the article is “critical thinking.” My concern, the take home message may have been something else.
I would like to offer five reasons to continue Blood Glucose Level (BGL) Point of Care Testing (POCT) for 911 (medical) responses as a standard checklist item for vital signs.
1. Data: Based on the data Dr. Bledsoe cites (from the paper published by Vihonen et al. that retrospectively looked at EMS encounters in Finland), the math used in the JEMS article is not accurate. While the author points to a total patient population of 707,417, only 15,158 patients were tested. Of those 15,158 patients, 3,856 (25.4%) had a blood glucose level (BGL) of <70mg/dL and 910 (6%) <54 mg/dL. We do not know the values for the remaining (707,417-15,158=692,259) because those patients were either not tested or their results were not recorded. Thus, those patient numbers should have been excluded in the denominator. The only number that should have been used for the denominator is 15,158.
2. Structure/Checklist: All too often, we (the EMS profession) are not as focused for our “ninth unknown rescue of the night” when it is 3am and we are summoned only because it appears local law enforcement has offered our potential patient either a trip to jail or the hospital. Nevertheless, it is a regimented structure (checklist) that helps us stay focused on the basics of a routine assessment to include BGL. Do you think Airline Pilots with tens of thousands of flight hours occasionally failed to set the flaps correctly before takeoff in the 1970s before the aviation industry embraced (as required by the FAA) checklists? Yes. Do you think surgeons with their years of training and experience would ever close a surgery while instruments or sponges were left inside the patient before reluctantly embracing checklists in the 1990s? Yes. Checklists help us avoid missing critical details.
3. Costs: Most devices (glucometers) used for POCT are either free or purchased at a nominal price, with the test strips having a cost of $0.50 to 1.00. Very few things we do in healthcare offers a better return on investment. Now, that does not stop some agencies from charging $50 for a POCT but that’s another story.
4. Past experiences: While it has been more than a few years since I worked as a paramedic or taught paramedic programs, I still stay actively engaged in EMS research. I have personal experiences where a routine BGL was the trigger for a more complete assessment that indicated hypo/hyperglycemia in patients who were otherwise asymptomatic for hypo/hyperglycemia. On a personal level, it was a routine football physical (where we expected to get a signature on a form and be on our way) that identified an asymptomatic family member with T1DM. While this was not an emergency episode, the 325 mg/dL finding and happenstance diagnosis most likely helped us avoid one.
5. Trends in medicine: According to the American Diabetes Association, approximately 10% of the US population had diabetes in 2015. According to Rowley et al., diabetes as a diagnosis in America is projected to increase by 54% by 2030. Diabetes is a public health crisis in America.
Full disclosure; I am not a physician. While I continue to be involved in research focused on pediatric care, burn injuries, and disaster medicine, I am a university professor in a business school teaching healthcare management. Furthermore, it has been many years since I actively worked as a paramedic taking care of patients so I do not pretend to suggest that my experiences and knowledge are greater than either that of the author or many of those who read JEMS and work in the EMS profession today. I do agree with the premise that all healthcare professionals need to continuously work to improve their critical thinking skills and there are times it is reasonable to not check a BGL or for that matter, other vital signs such as blood pressure (e.g., alert/oriented patient who is autistic). Regardless, I urge you to focus on the critical thinking part of this article and not use this as a means of dropping the BGL from your routine assessment checklist simply because you read an article in JEMS.
Randy Kearns, DHA, MSA, FACHE, FRSPH
Assistant Professor, College of Business Administration, University of New Orleans
Retired Clinical Assistant Professor, School of Medicine, University of North Carolina
Dr. Bledsoe’s response:
Thank you for your comments. It is always nice to hear from you, Randy. With regards to Finland paper, the informed decision not to test implies that testing was not indicated. Granted, there may have been some that were not tested and had hypoglycemia, but the study design did not reveal this. One-hundred percent blood glucose testing in any system is unreasonable. Regardless, the incidence of hypoglycemia in non-diabetics is extraordinarily uncommon.
I agree that checklists are good. We use them for high-risk situations and procedures like those you mention. Simple assessment tasks generally do not require checklists except for novice providers. The routine assessments discussed here are not complex.
With regard to costs, I agree with you for the most part (some departments use i-STATs® which are not inexpensive). However, many services use the glucose determination to bill an advanced life support charge.
While the past experience you describe is notable, the patients I refer to are, by definition, being transferred to a health care facility where more comprehensive testing will be performed. While we make note of prehospital blood glucose levels, we confirm with our own equipment (and the differences between the prehospital and hospital blood glucose levels are often significant). Further information is required to establish a diagnosis of diabetes.
With regard to trends you mention in medicine, patients need to be screened and tested in the hospital with calibrated equipment. Patients with possible diabetes we will get a hemoglobin A1C, ketones, and an electrolyte panel. Hypoglycemia is the biggest concern in patients with diagnosed diabetes. We often discharge T2DM patients with glucose levels in the 200-400s if they have follow-up care available. T1DM patients with hyperglycemia tend to develop DKA and are usually admitted. That said, setting up a community screening program and testing people is a good use of testing and a community benefit.
Bryan Bledsoe, DO, FACEP, FAAEM, EMT-P
Professor of Emergency Medicine, University of Nevada School of Medicine
Attending Emergency Physician, University Medical Center, Las Vegas