Overdiagnosis of conditions other than cancer is occurring, and researchers now have a way of estimating it.
Researchers from Bond University's Institute for Evidence-Based Healthcare (IEBH), Dr Sharon Sanders, Dr Mark Jones and Professor Paul Glasziou AO, and researchers from the University of Sydney, the University of Queensland, Monash University and The Australian National University have developed and applied a ‘Fair Umpire’ framework for detecting and estimating overdiagnosis in non-cancer conditions.
“We’ve known for a decade that cancer overdiagnosis is happening and how to measure it. Until now however, that hasn’t been possible for non-cancer conditions. Now we have a valid approach for evaluating whether overdiagnosis of non-cancer conditions is happening, and that opens up options for being able to mitigate it”, said Professor Paul Glasziou.
Overdiagnosis in non-cancer conditions can occur when a new diagnostic approach, or a new disease definition is adopted. Such changes generally lead to additional diagnoses, not fewer diagnoses, and some of the additional cases may be overdiagnosed.
“The Fair Umpire framework uses ‘Umpires’ – that is patient outcomes and tests (not people) to adjudicate the extra diagnoses. The different umpires are able to tell us, to different degrees, whether those additional cases are likely to be overdiagnosed and may also allow the calculation of estimates of overdiagnosis.” Dr Sharon Sanders, an Assistant Professor at IEBH said.
In the first global systematic review of research studies on overdiagnosis in non-cancer conditions, the researchers applied this framework to 132 studies that examined overdiagnosis in respiratory, cardiovascular, urogenital, musculoskeletal, mental health and other conditions. This review was published in the Journal of Clinical Epidemiology in January 2024.
“This review is the first to comprehensively assess overdiagnosis in non-cancer conditions using the new Fair Umpires framework”, Dr Sanders said.
“We found that almost one third of the studies did not use any Fair Umpire to adjudicate the extra diagnosis and, as such, provide limited evidence for overdiagnosis. The remaining studies did use a Fair Umpire, such as a clinical outcome or follow-up or additional test to assess the extra cases diagnosed, and in 32 of these studies it was possible to estimate the percentage of additional cases that were overdiagnosed”, she said.
Estimates of overdiagnosis included 43-45% of abdominal aortic aneurysms detected by screening, 27% of women diagnosed with diminished ovarian reserve using the 2011 diagnostic strategy (compared to the 2004 diagnostic strategy), 54% of children diagnosed with acute kidney injury by RIFLE criteria, and 77% of pregnant women diagnosed with oligohydramnios by colour doppler sonography.
Professor Katy Bell from the University of Sydney, who was lead author on the study first describing the Fair Umpires framework said, “this study is important because it shows the usefulness of the Fair Umpire framework for detecting and quantifying overdiagnosis of conditions other that cancer, which was not possible before.”
“Overdiagnosis is a particularly challenging and contentious area to talk about and mitigate. Being able to generate robust evidence about whether its occurring and how often is really important to make sure that diagnosing additional people is actually helping rather than harming them”, said Professor Glasziou.