There is plenty of proper science on this. Weight gain does not cause obstructive sleep apnea until you get into extremes (e.g. huge necked bodybuilders or people with so much fat on their chest that they physically struggle to move it to breathe). Sleep apnea makes it harder to lose weight and easier to gain weight. Having sleep apnea and being heavier can make sleep apnea worse. Losing weight quickly can make sleep apnea worse when you lose muscle mass along side fat (e.g. on ozempic).

There are plenty of tools for doctors to treat sleep apnea. The problem is that they refuse to use them. Many people on CPAP would benefit greatly from being on BiPAP instead, but doctors commonly refuse to prescribe it. Some cases of sleep apnea can be treated using positional therapy (typically side sleeping), but there's no prescription for that. Some cases can be solved by exercising throat muscles (with or without a didgeridoo), but there's no prescription for that either, and there are virtually zero speech/physical therapists who focus on that. There are some surgeries that can really benefit some patients, but most sleep labs and ENTs refuse to even to even perform a proper sleep endoscopy.

The idea people aren't doing 'proper science' on this is a spectacular level of oblivious. There's nothing in here that I haven't at some point had a discussion about with my sleep doc in the last 20 years (well, not specifically a didgeridoo, but circular breathing and other types of breath exercises, including 'straw breathing'). Yes, some people do loose weight when they get their apnea under control, among other health benefits. That also is not new, and it's not some sort of miracle insight that noone has considered.

I had textbook symptoms of sleep apnea with a BMI of 19, before I was diagnosed. The sleep tech told me bluntly that plenty of slim people, even children, develop sleep apnea.

Obesity increases the chance of developing sleep apnea, yes. But sleep apnea also increases the chance of becoming obese. It is not just a simple unidirectional cause and effect.