If you can, ask the people over there what they need more urgently! You deciding what's best for them is… typical white people behavior.

If you can't, well…

This is the result of grabbing the first reasonable-looking results off Google and throwing stuff together, where "reasonable-looking" is based on 2 months of intensely tracking the CoV development and ZERO prior experience.

Ventilators (that's the fairly easily quantifiable part)

• About 40-85% of people put on a ventilator die. (This reddit post has / links a bunch of estimates, but no clear numbers exist…)
• Besides ventilators, you need trained personnel, and various drugs (for sedation & relaxation – the patient shouldn't fight against the respirator.)
  • Basic training should not be a problem (lots of people/institutions are putting videos online), but lack of experience / risk of errors will probably further lower the survival chances.
  • You still need lots of personnel – ventilation is high-maintenance.
• Prices seem to be about 500 per, so you'd get about 40 ventilators.
• For time on a ventilator I'd say average about 10 days (some die quicker, some need 3 weeks to recover enough)

So assuming personnel/drugs/etc. are absolutely no problem, 40 ventilators will support about 120 people per month, which will save about 30-70 lives. Assuming a peak duration of 3 months (completely made up) that's about 100-200 people saved?

PPE (This is very soft / fuzzy… no clue.)

• Nitrile gloves seem to cost about .20 per pair
• simple masks are (were?) also about .20 per (if you can get some…)

so if you don't reuse equipment at all, basic equip is about .40 per set or you can have 50K reasonably safe patient interactions with the $20K.

So how many lives does that save…? That depends on how/where the PPE is used. (Solely to protect medical personnel from known-infected people, or when interacting with general people of unknown status / protecting both sides from each other, or also filling in other shortages in the normal "background noise"?) No clue, so let's just say 10% (completely made up number) of those interactions would have resulted in an infection. That would result in 5K prevented infections, which at 5% of them requiring ventilator/ICU would prevent about 250 critical cases that (assuming no ventilators are available) would (probably) die. (Even with ventilators that's about 50-180 deaths?)

So from my estimates, they come out at roughly the same level, with PPE probably being better, but variance for PPE is much higher. (Could save a lot more or a lot less, and if not being down with the virus for several weeks also counts as fractional lives it's probably a lot higher.) Still, could go either way – so, if you can, ask.