I think they mean started in terms of treating patients with TB, as a physician rather than this particular patient? But maybe you’re right in which case I’d guess they’d be out cured much sooner.

Well, I don't have a supplier for this medication currently, and a study size of 109 is not large. The results are very promising but with TB, only certain medications work in certain combinations and you don't get to find out if you treated it wrong until sometimes years later when they relapse.

If this particular patient asks about this study, and I won't be at all surprised if they do because this is a major article in a major paper and they're pretty well informed about their condition, I would probably tell them that while it looks very compelling I don't want to do this incorrectly and right now the best evidence is the WHO guideline as currently followed. Without going into detail, this patient is already on two of the three meds as it is.

Couldn't say definitively, but given that the biggest risks for resistance are monotherapy and intermittent therapy, shorter, less complex courses help to reduce the effective occurrence of both. Currently my MDR and XDR patients have about an 18+ month time frame; slashing the treatment time potentially to a third is a huge win, and we have robust directly-observed therapy infrastructure to ensure doses are actually taken. Unfortunately this last is not guaranteed in the developing world.

The article covers that, in brief:

Tuberculosis germs burrow deep into the lungs and barricade themselves inside clumps of dead cells. Breaking those nodules apart and killing all the bacteria inside requires taking drugs for months.