I note the authors' definition of overdiagnosis includes "people without symptoms ... diagnosed with a disease that ultimately will not cause them to experience symptoms or early death".

In my limited experience I believe that, despite good end of life care, overdiagnosising may occur in that sub-population that are symptomatic and do have a separate condition for which there is treatment, but who are already at a point where treatment would not significantly improve life expectancy or quality of life.

I have been called to take blood from octogenerian patients with cardiorenal syndrome and painful swollen arms. The phlebotomist had failed, the nurses had failed, and I am sorry that I failed too. In the midst of rushing between wards, I took some time to talk to the patients about why they didn't want any more needles and blood tests, and I realised I might have crossed the fine line between doing good and doing harm.

If the cause to prevent overdiagnosis includes a focus on those who are not necessarily healthy but who would not benefit from another new diagnosis, could it serve as a useful evidence-based approach to improving quality of end of life care?

Competing interests: None declared