In many cases these doctors aren't practicing physicians, iirc the reviewer in the article hadn't practiced for 25 years. They just need someone with an MD to sign off on the denials.
That's not the same as what doctors and nurses do.
They get a bunch of test results and a first opinion (from doctor) then they issue a new (second) opinion of no treatment. Doctors give out second opinions all the time; it's a thing patients do when they want to make sure the first one is correct.
Doctors have direct contact with the patient and are pursuing leads via testing and originating a treatment plan.
Ergo, "what might work"
A medical reviewer is limited to generated documentation only (though can request more) and is then comparing that to relevant regulations, standards of care, and reasonable/necessary tests.
Additionally, they are a technical expert in recognizing fraud that may be hidden in individually-reasonable, unreasonable-in-aggregate cases.
Ergo, "what can be justified"
Here's the CMS-version explainer, for context: https://www.cms.gov/data-research/monitoring-programs/medica...
Generally, criminalizing malpractice is a counter-inventive to actual system improvement. See doctors. They just add malpractice insurance, and the cost is added to everyone.
If the real concern is that medical reviewers aren't fairly reviewing claims (which afaik, they generally are, contingent on documentation being available), then the supply side should be addressed -- mandate a specific, minimum review time per claim, a standard appeals chain that all insurers are subject to, and staff a centralized, independent (probably under CMS) final appeal arbiter.
Or is there a possibility that they could lead nowhere?