80% of antibiotic use in the USA goes to farm animals, not people. We could therefore significantly reduce the amount of antibiotics in our environment without affecting human lives.
You could make another very significant reduction by keeping doctors from prescribing antibiotics so often to placate patients who really don't need them.
This would go a long way towards reducing bacterial resistance without changing what treatments are available for truly sick people.
Good luck with that. This would mean regulation, and that would mean going up against corporate farms and their lobbyists as well as the Senators and Congressmen they have in their pockets. America's legislative branch is actively hostile to science and anything that could impact the bottom line of their campaign donors.
I guess the alternative is to hand wave that us scientists can magically solve the decades-building problem that big-agriculture and doctors caused. Sure, we'll get right on it, but there needs to be a plan for the time between when research starts and when a new group of antibiotics is tested and ready for the market. Staged rotating/phasing out certain antibiotics so that they will be usable again after a period of time would ensure that there's something left in reserve for people who actually need it.
Tuberculosis is a prime example of why this is such an intractable problem. 1 in 3 people globally are infected with the TB bacterium. That's >2 billion people!
It is usually dormant, but there's a 10% chance it will flare up into the active - and deadly - state in a healthy individual's lifetime, so it really needs to be treated. It is the top infectious disease killer every year (>1.5 million people, including 140,000 children).
The current treatment regime requires daily doses of multiple kinds of antibiotics for 9 months, and that assumes you're lucky enough to have a strain that is vulnerable to first line drugs. In practice many treatment regimes require multiple antibiotics administered at once, and can take up to 24 months to clear from the body. Second line drugs have severe side effects.
Every time someone misses a dose, they risk creating a resistant strain. In fact, even if you're taking it regularly, there's a chance it will still develop into a resistant strain.
There is simply no way to ensure all of the 2 billion infected people around the world, mostly in low and middle income countries, will (1) be identified, (2) have access to these drugs, (3) consistently take them for 9-24 months even if faced with severe side effects, and (4) be monitored by a professional that can decide when to move to a second or third round of treatment.
A certain percentage of the population will never be treated, abandon treatment too early, or simply have terrible luck and not fully respond to treatment. You just can't identify and quarantine them all. Besides, even if you did, where would you house them? For how long, the 24 months of treatment? The rest of their lives?
It's also important to remember the human element here. We can't cycle/phase out antibiotics, even if you did manage to get every doctor in all 190+ countries of the world to agree to do so. What would you do if you or your baby needed treatment, but a doctor told you "no, we won't prescribe the most effective treatment because we're worried in 30 years it won't be effective anymore." Without effective treatment, an active case of TB has a 45% chance to lead to death. It's also in everyone's interest that that infected person be cured asap.
Finally, it's highly doubtful draconian containment or drug cycling restrictions would even work. Drug resistant genes are already in the natural environment. Life finds a way. In 2012, doctors found a strain of TB in India that is entirely resistant to =all= known treatment. They found 12 cases in one hospital alone. Most facilities don't know how to identify it, so it's spreading out there, right now.
Quarantine and treatment can slow the progression of resistance, but it simply can't be reversed. 3.5% of new cases every year are already resistant to =multiple= drugs. We will need to develop an entirely new model over the next century to eliminate these bacterial infections.
Different antibiotics have different pathways, much like barbituates and amphetamines have different pathways, despite both getting you high. Perhaps if we use only pathway "A" for a decade, bacteria will lose resistance on pathway "B" (and vica versa)?
In order for us to "stop the arms race for stronger and stronger antibiotics", we have to let people infected with resistant bacteria die. In droves. You would be all for this strategy until it's you, or your spouse or child sitting there being eaten alive by MRSA. You could tell your toddler that she's not dying for no reason, and that a few years from now the bacteria won't be as deadly. That'll make her feel better.
Do you really think that this is even remotely realistic in any democratic society?