[Correspondence] Factors that can make overall survival unreliable as a clinical trial outcome – Authors' reply

We were surprised to read the letter from David J Stewart and colleagues in response to our Comment because they dismiss overall survival—the endpoint that patients have consistently reported as most meaningful1—as difficult and unreliable. We agree that measuring overall survival is difficult, but that cannot be an excuse to not measure it. We disagree that overall survival is unreliable.…