The United States Preventive Services Task Force (USPSTF) recently updated its clinical practice recommendation regarding screening for type 2 diabetes in children and adolescents, coming up with an “I” statement of insufficient evidence due to lack of studies demonstrating that interventions for screen-detected type 2 diabetes improve health outcomes.
The American Diabetes Association (ADA) also updated their clinical guidance in 2022, but this group offers up a Grade A/B recommendation that we should consider screening in children who are at least 10 years old or pubertal and who have risk factors.
This happens all the time — differing clinical practice recommendations from different professional organizations. Which recommendation should we follow? Who do we believe? And why?
First, we need to clarify the scope of the recommendations. In this case, the USPSTF is concluding on the general population of all children and adolescents, whereas the ADA is suggesting screening in a subpopulation who are overweight and have other risk factors. So, it’s a little like comparing apples to apple slices.
It is also important to consider sources of potential bias, which can be intellectual, financial, or philosophical. The USPSTF is an independent, government-funded panel of volunteer, nationally recognized experts in prevention and evidence-based medicine. These experts have no significant conflicts of interest and engage in a process of systematic appraisal of data in a transparent and reproducible way. The ADA is a private advocacy group funded in part by the pharmaceutical industry, so there is potential for a financial incentive when more patients are diagnosed with and treated for diabetes.
Philosophical differences among professional organizations tend to look like a bias towards action or inaction when data is lacking. In general, specialty organizations tend to err on the side of “doing something” whereas primary care groups tend to “first do no harm” when there is not enough evidence to draw a firm conclusion.
Screening to identify disease earlier than it would manifest naturally is not always beneficial. Some diseases never go on to cause clinically important harms, and some will spontaneously improve. Prostate cancer is a good example of when simply identifying a disease may have more harms than benefits. When it comes to type 2 diabetes, we still don’t know whether identifying children by screening leads to better long-term clinical outcomes. But we do know that overdiagnosis can lead to overtreatment which can result in harm.
An “I” statement of insufficient evidence from the USPSTF can be mistaken as a weaker recommendation or an “I don’t know” statement. It’s not, though. An “I” statement offers transparency in appraisal of the evidence, or lack thereof, and often stands as more valid and trustworthy than recommendations that use stronger language but are based on lower quality evidence or expert opinion.
Practice Point: Interventions for screen-detected type 2 diabetes in children and adolescents have not [yet?] been shown to improve health outcomes. Current evidence is insufficient to recommend screening.
EBM Pearl: An “I” statement from the USPSTF does not stand for “I don’t know.” Take the time to critically consider more strongly worded recommendations from other groups when the USPSTF concludes that the evidence is insufficient.
For more information, see the topic Diabetes Mellitus Type 2 Screening in DynaMed.
