What we know and don’t know about overdiagnosis

As the body of research on overuse grows larger each year, it is increasingly important to evaluate the research and identify the gaps. This week, six researchers from The Netherlands published the results of an ambitious project in The BMJ – the first scoping review on overdiagnosis research. They examined 1,851 articles  for patterns and discrepancies in the definitions of overdiagnosis and methodology. They also compared how prevalent overdiagnosis research was across clinical fields and countries. 

Here are a few key points from the review:

  • Studies on overdiagnosis in Oncology made up half of the studies in the review, the largest proportion of any clinical field by far. Mental health (9%), infectious diseases (8%), and cardiovascular diseases (6%) were the next highest represented fields. Despite the dominance of oncology in overdiagnosis research, the authors found that there were some studies on overdiagnosis in every specialty. 
  • There isn’t one standard definition for overdiagnosis, which can make it difficult to compare the results of studies. Does overdiagnosis mean a diagnosis made without indicated symptoms? Does it mean a diagnosis of a condition that’s actually within the scope of human variation? Or does it mean when someone is diagnosed with a condition that will never go on to harm them? Depending on what type of overdiagnosis you are trying to measure, there could be many different definitions.
  • Although the studies all used the term “overdiagnosis,” the researchers identified many distinct types of overdiagnosis, for which they recommend additional classification. For example, one could classify diagnosing someone with a harmless cancer as “maldetection,” which would be detecting harmful cancer where there is none. Another type is “misclassification,” when a patient’s underlying condition (blood pressure level, eg) hasn’t changed but the threshold for disease changes (such as the recent adjustment of blood pressure guidelines). Both are overdiagnosis, but there are different drivers and interventions for each.
  • The United States had the highest representation of authors on overdiagnosis studies, with 38.7% of authors from the US. Next was the UK with 10.9% and Australia with 6%. However, the geographic scope of overdiagnosis research is quite wide; there were authors represented from 65 different countries across the globe. The map below from Dr. Jessica Otte of Less is More Medicine illustrations how widespread this research topic has grown:
Source: Dr. Jessica Otte, Less is More Medicine

Overdiagnosis is being discussed in almost all clinical specialties all over the world, but we’re still missing a standard definition and language to compare evidence across location and field. This scoping review helps provide a path forward for future research on overdiagnosis.