Author’s reply to Braithwaite

I’m grateful to Braithwaite for giving me the opportunity to bang on about this a bit more.1 2 He misunderstands—the point is that Mitchell and colleagues’ study cannot account for rising prescription rates because they aggregated data across more than two decades.3 These authors do not chart a change in diagnostic practice, so the findings cannot support the idea that a loosening of diagnostic categories over time accounts for increased prescribing as Godlee suggests.4

The crucial point here, as Mitchell and colleagues point out,1 …