BlackSpinner wrote: Well that doesn't say much for your math training.
Thank you for not making your attack personal. But my math teachers would like to say to you, "a pox on you." I would never say that myself, of course. I respect you too much.
BlackSpinner wrote: If out of 100 people 75 are diagnosed with garden variety OSA
So, good math is where you just make up your own statistics out of thin air? Cool. I did not know that. I learn so MUCH on this forum.
BlackSpinner wrote: at $300 that is $22,500 while 75 at $2000 (cheap sleep lab) is $150,000 for difference of $147,750. For the rest of the 25 people who fail the home test, because it is designed to catch garden variety OSA and NOT to rule out sleep apnea you now run them through the whole sleep lab experience but for the 75 people with valid diagnosis you have saved $147,750. Which means the people who have problems putting on their little belt and finger clip - they end up in the 25% group.
And 75% is low balling it.
And if you are really challenged arithmetically that means for 100,000 people you save $147,750,000 which is a nice number.
Very imaginative.
But, to my eyes you did not address one of my points to the wording you quoted before your little exercise of imagination.
Many, many patients don't qualify for HST if you use the AASM standard, so they are going to get a PSG. So now we are only talking the number of patients that DO meet the AASM criteria, if we are keeping up standards. That lessens the suppposed savings before we even start.
Now for the other issues with the ones who do meet the criteria:
How many retests of the HST are needed, on average, even among the age group and literacy group and aptitude group served, because of test issues that render the data, miniscule though it is, inconclusive? How much time and money is wasted on that? After several retests for one patient, all the "savings" from the inferior test melt away quickly.
How many patients test positive on the HST but eventually should have an attended diagnostic PSG anyway? Maybe they don't get one. But how many SHOULD have it? Because if they should, but don't, that is directly opposed to the interests of that patient in a VERY significant way.
How many patients test negative but get an attended PSG anyway? Don't ALL of them, since they were all suspected of having OSA? If so, the HST was an awfully expensive pretest to the real test and one that was a complete waste of time, since it didn't rule out or rule in anything in its "results." In that case, the HST represents an ADDED expense, not a savings of ANY sort.
And then, what if the patient is denied a PSG because of negative results of the HST, and that patient goes on to develop serious health problems, the payments for which come out of your pocket and can't be used toward your knee replacements, to use your example. How did you factor those patients into your math? You didn't? Oh, sorry. My math teachers would like to talk to you about that after class.
If only a relatively small subset of patients qualify for HST, it is never good math to make it sound as if all who are suspected of OSA can take one instead of PSG to save money. And trying to compare them one-to-one the way you did looks good on paper, but doesn't always work that way in the real world, I don't think. That's why imaginary math is fine. Insurers get fooled by that "logic" all the time. But the points I made were an
attempt to point out the limitations of the theories about money saved. I have nothing against saving money when it doesn't sacrifice patient care. But HST is inferior to PSG by far. If no PSG is needed, then, in my opinion, no HST is needed either, since there are better ways to diagnose and treat patients than using HSTs to prove to insurance what the doc and patient already know. And when the criteria for HST are met, doc and patient usually already know.
That said, I agree that there are some circumstances in which HST is very valuable. They are one of many tools that should be available to sleep docs. But educated sleep docs should be deciding when they are useful by using judgment that is not unduly influenced by the bean counters, in my opinion. Cheaper isn't always better. Sometimes, cheaper means you get what you pay for. In this case, a test that proves next to nothing is very expensive compared to a test that actually proves something.
In my opinion.
But hey, I only made C's in math. That's why your words hurt my feelings.