It would depend on the scale being used. It could be either way...

I had a psychiatrist once who developed his own rating scale based on a series of questions. I think the scale was low = less depression symptoms reported and high = increase of depression symptoms reported. But I believe that was an ongoing measurement, one that compared one reporting to the next, and the scale was relevant to each patient individually.

Comparing the results over time was helpful to the psychiatrist and the patient, to see how different medications helped or didn't help, for example.