Common condition is often misdiagnosed, with dangerous results

Naveed Saleh, MD, MS, for MDLinx | June 05, 2019

More than one-third of adults over age 30 who receive an initial diagnosis of type 2 diabetes may actually have type 1, according to the results of a study published in Diabetologia. Specifically, researchers found that misidentification of late-onset type 1 diabetes is fairly common, as 38% of patients with type 1 disease, occurring after age 30, were initially treated for type 2—a misdiagnosis that may persist even years later and lead to adverse health outcomes without proper treatment.


Many adults may appear to have type 2 diabetes, but they might actually have type 1.

Indeed, more than a decade after misdiagnosis, half of patients were still being treated as though they had type 2 diabetes rather than type 1.

“While people with type 2 diabetes may eventually need insulin, their treatment and education is very different from type 1. If people with type 1 diabetes don’t receive insulin, they can develop very high blood glucose, and may develop a life-threatening condition called ketoacidosis. This means having the right diagnosis is vitally important even if insulin treatment has already been started,” noted first author Nicholas J. Thomas, MRCP, Institute of Biomedical and Clinical Science, University of Exeter Medical School, Exeter, United Kingdom.

Yet, distinguishing type 1 from type 2 diabetes in later life can be difficult, with 42% of type 1 diabetes cases occurring in adulthood after age 30, according to prior research.

“It is very difficult to diagnose type 1 diabetes in older adults, as most people of this age will have type 2, even if they are thin,” explained senior author Angus G. Jones, PhD, Institute of Biomedical and Clinical Science, University of Exeter Medical School.

With this in mind, Drs. Thomas and Jones and colleagues assessed the prevalence and characteristics of type 1 diabetes, defined by rapid insulin requirement (within 3 years of diagnosis) and severe endogenous insulin deficiency (non-fasting C-peptide < 200 pmol/l), in 583 participants with insulin-treated diabetes diagnosed after the age of 30, from the Diabetes Alliance for Research in England (DARE) population cohort. They also assessed whether patients with type 1 disease were correctly identified and managed in clinical practice.

The characteristics identified from the experimental cohort were compared with those from 220 subjects with youth-onset (< 30 years of age) type 1 diabetes characterized by severe insulin deficiency, as well as participants with retained endogenous insulin secretion (> 600 pmol/l).

In total, 21% of participants with insulin-treated diabetes diagnosed after 30 years of age met criteria for type 1 diabetes. Among these participants, 38% did not receive insulin at diagnosis, and 47% of those who delayed insulin treatment self-reported type 2 diabetes.

Severe endogenous insulin deficiency was highly predicted by rapid insulin requirement. In sum, 85% of participants needed insulin within 1 year of diagnosis. Furthermore, 47% of all participants who were initially treated without insulin and went on to need insulin treatment within 3 years of diagnosis exhibited severe endogenous insulin deficiency.

Furthermore, the investigators found that participants with late-onset type 1 diabetes, defined by the development of severe insulin deficiency, exhibited clinical characteristics similar to those with youth-onset type 1 diabetes. However, those with late-onset type 1 diabetes demonstrated modestly lower type 1 diabetes genetic risk scores, increased islet autoantibody prevalence, and lower odds of identifying as having type 1 diabetes.

Drs. Thomas and Jones and colleagues also found that classical clinical criteria cannot reliably distinguish individuals with late-onset type 1 diabetes, thus resulting in severe insulin deficiency. In addition, they noted that 30% of participants who received diagnoses at > 30 years of age and given insulin at diagnosis had type 2 diabetes per C-peptide criteria, with only 25% of these participants self-reporting a diagnosis of type 1 diabetes and only 59% receiving oral glucose-lowering therapy at a median diabetes duration of 10 years.

Overall, study findings on the clinical features of late-onset vs young-onset type 1 diabetes were in line with results from recent novel population-based genetic stratification analyses. Notably, 89% of patients with genetically defined type 1 diabetes required insulin treatment within 1 year of diagnosis—a finding similar to the 85% reported in this study using a C-peptide-based definition.

The authors acknowledged that their study had some limitations, including its cross-sectional design within a homogenous, geographically restricted, English population.

Nevertheless, Drs. Thomas and Jones and coauthors concluded: “These results have clear implications for clinical practice. They show that type 1 diabetes leading to endogenous insulin deficiency is common in later life but is difficult to identify…Our results suggest that if patients are treated as having type 2 diabetes but progress to insulin within 3 years of diagnosis, clinicians should reassess the underlying diagnosis and strongly consider biomarker testing.”

Test your knowledge on Diabetes by playing Smartest Doc!