> Though few patients appeal coverage denials generally, when UnitedHealth members appeal denials based on nH Predict estimates—through internal appeals processes or through the federal Administrative Law Judge proceedings—over 90 percent of the denials are reversed, the lawsuit claims. This makes it obvious that the algorithm is wrongly denying coverage, it argues.

The '90% error rate' in the title sounded like a pretty serious twisting of some numbers so I scanned the article for where it's quoted from ^, which... makes a mockery of the title.

In case it's not obvious and to save parsing that, 'few people appeal' (which makes sense given the internal appeals process or Administrative Law Judge proceedings). Then, /of those cases/ 90% are overturned.

- If 50 / 10,000 cases and 45 were overturned or 500 / 100,000 cases appealed and 450 were overturned... would all meet this '90% error' mark.

- Similarly, 'overturned != error', because it may be as simple as a policy on United's side to not fight super hard and allow overturns (for any number of reasons, including just saving costs on the proceedings, it needn't be altruistic).

Did Ars used to be better than this, or were they always like this on some issues?

The /business/ side of medical insurance isn't really a topic that I know enough to weigh in on, but hopefully people can read the above sentence and parse what it's actually saying for themselves before reading the article.

Irrespective of how I/we feel about the outcomes, this headline is awful.