Interview with Dr. Irl Hirsch on the Problem of Misdiagnosis

Recently, I had the great pleasure of interviewing Dr. Irl Hirsch about the problem of misdiagnosis—people with Type 1 diabetes who are misdiagnosed as having Type 2 diabetes.  I had heard Dr. Hirsch speak at the Taking Control of Your Diabetes (TCOYD) ONE Conference in San Diego in June 2017, and over the years I have read a number of his papers and know he is a great advocate for those with adult-onset Type 1 diabetes.  Dr. Hirsch is a preeminent endocrinologist, diabetes researcher, author of numerous articles and books on diabetes, and a person with Type 1 diabetes himself (since he was 6 years old). He is a professor of medicine at the University of Washington School of Medicine and sees patients (the majority with Type 1 diabetes) at the University of Washington Endocrine and Diabetes Care Center.  Oh and did I mention he is brilliant and has a great sense of humor? When the opportunity to sit down with Dr. Hirsch for a one-on-one conversation presented itself, I took it.  What came next was a whirlwind 30 minutes of conversation that educated, validated, and reinforced my mission to change the way the medical community diagnoses adult-onset Type 1 diabetes. 

There were two key themes that Dr. Hirsch reiterated throughout our conversation:  first, adult-onset Type 1 diabetes is difficult to correctly diagnose and second, physicians do not have the necessary time to spend with patients to achieve good outcomes.

First let’s talk about why it is hard to correctly diagnose adult-onset Type 1 diabetes.  Dr. Hirsch sees misdiagnosis as an epidemic—in Dr. Hirsch’s busy clinic, they see approximately one new person per week who is misdiagnosed (diagnosed as Type 2 when the person actually has Type 1).  From his viewpoint, there is such a heterogeneous phenotype in adult-onset Type 1 diabetes, from a classic phenotype of rapid onset (rapid weight loss, Caucasian, normal BMI, polyuria, polydipsia, polyphagia) to a very slowly progressive Type 1 phenotype, that makes it more challenging to correctly diagnose.  As Dr. Hirsch repeatedly said, the presentation of Type 1 diabetes in adults is so variable.

In my opinion, one of the fundamental problems is that medical and nursing schools do not teach the most current information about Type 1 diabetes, and they still perpetuate the myth that Type 1 diabetes is a childhood/young adult disease.  What chance do we ever have of changing how doctors think if the information they get in school is out-of-date and/or flawed?  Of course, at the University of Washington School of Medicine where Dr. Hirsch teaches, they train medical students that new-onset Type 1 diabetes is seen in all ages and has a variable presentation in adults—they are working to make sure future doctors know the range of Type 1 diabetes presentation and how to correctly diagnose and treat it.  I am reminded that change comes one step (or School of Medicine/medical textbook) at a time.

As technology improves, it brings with it better diagnostics but those advancements can be years or even decades to fruition.  Dr. Hirsch mentioned the United Kingdom Prospective Diabetes Study (UKPDS) that began in 1977 and he shared something about the research which reminded me of an old saying:  if only we knew then what we know now.  Some years after the UKPDS commenced, the researchers went back and performed autoantibody testing on the subjects’ stored specimens.  They determined that about 10% of people who were diagnosed with Type 2 diabetes were autoantibody positive and in fact had Type 1 autoimmune diabetes.  They also found that the autoantibody positive people tended to be younger, thinner, and Caucasian.  In the UKPDS, the average number of years before these Type 1s were put on insulin was six years, but they were not put on insulin until their fasting blood glucose was 270 mg/dl.  That is over 2,000 days of being incorrectly diagnosed and incorrectly treated because technology hadn’t advanced to the point it is today.  When I asked Dr. Hirsch why all Type 2 trials today do not use autoantibody testing to eliminate Type 1s from the trial, he reminded me that autoantibodies often go away over time.  So technology is getting better, but we aren’t quite there yet.

Even if we get to the point where the technology exists to diagnose every disease, every time with perfect accuracy, we will still be dependent on physicians to interpret the results, develop a treatment plan, and administer care.  This leads us to the second theme of my conversation with Dr Hirsch:  physicians don’t have enough time to spend with their patients to achieve good outcomes. 

In Dr. Hirsch’s words, the United States’ medical system is not set up to provide excellent care for people with diabetes, both the newly diagnosed and those with long-standing diabetes.  This is one of those topics Dr. Hirsch writes about in his annual “Dr. Irl Hirsch’s Rant.”  Google it, read one or more, and you understand what a passionate advocate Dr. Hirsch is for better medical systems and care.  I asked Dr. Hirsch why all patients with new-onset diabetes aren’t tested for autoantibodies, to determine if the patient has Type 1 diabetes, and he said largely because it is expensive and controlled by insurance.  Dr. Hirsch emphasized that in the primary care setting, adult-onset Type 1 diabetes is not on the physician’s radar or in their bandwidth.  They see so many Type 2s and have so little time.

Dr. Hirsch sees many areas for improvement.  He says, “These days, 95% of doctors have employers and thus metrics that they must achieve.  They see patients every 15 minutes, and don’t have the time and infrastructure to teach patients how to use mealtime insulin.”  To further compound this problem, most clinics don’t have a Certified Diabetes Educator (CDE), which is one of the ways that people with new-onset Type 1 can learn the basics.  At a recent Endocrine Society meeting, Dr. Hirsch did an informal survey, and he found that 1/3 of endos have only 15 minutes with diabetes patients, and 1/3 have only 20 minutes with diabetes patients.  This means that only 1 out of 3 of these physicians will potentially have enough time to train a new patient on the use of insulin, or download an established patient’s pump and/or continuous glucose monitor (CGM) data and analyze it. 

Compounding the problem of limited time, Dr. Hirsch also described “doctor and patient barriers to insulin use.”  Many physicians won’t start a patient on insulin even if a patient has an A1C of 10 with classic symptoms.  According to Dr. Hirsch, a fundamental problem is that these doctors are not taught how to use mealtime insulin, are terrified of mealtime insulin, and thus they avoid it as long as possible.  If there were more time and support for the training, for example having a CDE on staff, perhaps this barrier could be overcome.  In some ways, the “patient barrier” is even harder to overcome because it can be based on cultural biases, fear, or shame—all powerful forces and obstacles.  Another patient barrier is the fact that people in certain jobs, such as commercial pilot or commercial truck driver, will lose their jobs if they must go on exogenous insulin.

In the end, 30 minutes went by in the blink of an eye.  Dr. Hirsch had to get back to his busy practice where he will again find this week’s misdiagnosed patient, change their diagnosis, change their care, and ultimately, improve their quality of life.  I am forever grateful for physicians and advocates like Dr Hirsch.   We didn’t solve the problem of misdiagnosis but we did have a conversation that gives me hope that one day technology, knowledge, and time will align and we will finally be able to correctly diagnosis and treat all people with Type 1 diabetes.