The ethics of bracing (and PT) with a Scoliscore <41

[hdugger]

But what is this other method? Isn't it reading and comparing serial radiographs and getting the answer AFTER THE FACT?

I don't believe so. They were comparing predictive tests.

I don't think this other test was sensitive, enough, btw, so I'm not comparing its efficacy. I'm only mentioning it because they thought the *goal* of the other test was inferior to the goal of predicting surgery.

No and it isn't the only thing that matters to surgeons. The problem is there is nothing available at the present time to help on any other front.

They specifically dismissed another measure because it didn't focus on surgery, which they regarded as being the important thing. That's a surgical focus which doesn't really take in the broad scope of what matters to patients.

The issue is you are assuming they don't care about certain things for reasons OTHER than they simply have nothing to offer patients to help. All the caring in the world about keeping sub-surgical curves even more sub-surgical isn't going to produce an effective conservative treatment out of thin air. Surgeons do what they can.

If conservative treatment doesn't work to keep kids under 30 degrees, then it equally doesn't work to keep children from surgery, right? It either does something or it doesn't. If that's the case, then there's no reason to pick one endpoint over another based on the treatment they can offer.

[Pooka1]

If conservative treatment doesn't work to keep kids under 30 degrees, then it equally doesn't work to keep children from surgery, right?

No that doesn't follow. It's an easier target to keep sub-surgical curves even more sub-surgical than it is to keep surgical curves sub-surgical it would seem.

It could be that a re-crunching of the brace data to look at this specifically or designing better bracing studies going forward will show that for genetically subsurgical cases, bracing MAY keep sub-surgical curves even more sub-surgical. The main problem they will be up against is the measurement error apparently in that smaller angle range. So the data will be noisier.

It either does something or it doesn't. If that's the case, then there's no reason to pick one endpoint over another based on the treatment they can offer.

Well, given the UNKNOWN fraction of people who are <50* at maturity who will go on to need fusion for any reason, and the KNOWN fraction >50* who need fusion, the endpoint of sugery is the most definitive and defensible end point. For all we know, the vast majority of people <40* at maturity will never need fusion for any reason. We know for a fact some do but we don't know the percentage of the total.

[hdugger]

No that doesn't follow. It's an easier target to keep sub-surgical curves even more sub-surgical than it is to keep surgical curves sub-surgical it would seem.

I think you're furthering my argument . If conservative treatment is *more* effective in keeping small curves small, then isn't the under 30 crowd what our doctors should be focussing on? Why emphasize keeping kids from surgery, if that's the thing they actually *can't* do?

[Pooka1]

I think you're furthering my argument . If conservative treatment is *more* effective in keeping small curves small, then isn't the under 30 crowd what our doctors should be focussing on? Why emphasize keeping kids from surgery, if that's the thing they actually *can't* do?

Well firstly, that is a big "if." There is no evidence that any conservative treatment can keep small curves small. I am just surmising that it should be easier to keep less susceptible, smaller curves smaller than more susceptible curves from becoming larger. Ever showing the former is going to be daunting because the researchers are up against formidable impedicments including measurement precision. Also some large percentage of these smaller curves will never need fusion for any reason. So they are looking for a small signal against a huge background. And given some perhaps large fraction of these folks <40* will never need fusion for any reason, the ethics of bracing has to be considered and these studies might never be done from that standpoint (just to bring it back to the original thread issue).

Secondly, not all surgons admit they can't keep surgical curves subsurgical with conservative treatments. So avoiding surgery is and should be the the major focus from a practical standpoint and a percentages argument.

[Ballet Mom]

http://adjust2it.wordpress.com/2010/...oliosis-brace/



A regression of 14*.

Also, there were 697 patients in the study and 75%, 24% and 1% were low, medium and high risk, respectively. Therefore:

Low - ~523
Medium - ~167
High - ~70

The girl used in the example should not have been used. A regression is obviously not going to be a typical experience for most of the people using the Scoliscore test and therefore can be classified as marketing and hype for the story.

I find it interesting that she has a score of 11. The girl in the story in the LA Times that I posted yesterday, had a score of 3. It appears her curve is not genetic at all and is due to needing a lift in her shoe.

It appears that only scores between 1 and 10 might actually be beneficial:

Ward said that a score between 1 and 10 shows no high risk factors for severe progression. "So far, we haven't seen any of those children develop a severe curve," he said. "Most of them don't even develop a progressive curve

You have to wonder how many patients score that low, and whether such an expensive test for the majority of the scoliosis population is worthwhile to bounce a few kids out of their braces.

Also they stated the following:


"Marissa is among the 2 percent to 4 percent of all youngsters older than 10 diagnosed with AIS, including about 30,000 who require braces and 18,000 who need spinal fusion surgery each year. "

This seems to be saying that 60% of kids who wear braces go on to have fusion. This is not consistent with the claim that only about 20% of kids can be considered bracing failures. Can anyone verify the numbers in the quoted statement?

There are 100,000 scoliosis diagnoses annually. 10% are juvenile cases, therefore less than the ten years old they're stating. 5% are infantile cases, also less than ten years stated. Adult scoliosis is apparently rare, so we'll ignore that. So basically, there are roughly 85,000 AIS cases annually. 18,000 fusion surgeries is a percentage of 21%. The small difference is probably due to the rapid progression cases that are hard to react to in time, and the unnoticed cases probably due to having a balanced double curve that is hard to identify until it's too late for bracing, and the small number of adult scoliosis cases, etc.

[Ballet Mom]

Originally Posted by Pooka1
Is he any different than the other surgeons giving the tests to their patients? Are they all "partners" in the research?

According to his site, he partnered with them during the development of the test, not just in using the test after it was developed.

The article is stating that he has a financial interest in the success of the Scoliscore test. You have to give the author credit for that.

Axial Biotech is a venture-capital financed company. The venture capitalists providing the financing to develop this test and company expect a big payout at sometime in return for providing this financing.

These privately-held venture backed companies don't have cash to pay for services so they typically use options, warrants, convertible securities, etc. to pay for services. Once the company is taken public, insiders can become phenomally wealthy due to ownership of these investment options. The venture capitalists will be pressing HARD to take this company public so they can get an enormous return on investment.

I am truly thankful that my daughter is just about through her bracing journey and isn't going to be used as a test subject as they roll out this test to the general population.

[hdugger]

I'm likewise suspicious, but I lean towards being suspicious by nature.

I would very much like to hear an evaluation of the test by an impartial SRS surgeon. I know Hart isn't signed up to administer it yet - maybe I'll ask him.

[skevimc]

There are 100,000 scoliosis diagnoses annually. 10% are juvenile cases, therefore less than the ten years old they're stating. 5% are infantile cases, also less than ten years stated. Adult scoliosis is apparently rare, so we'll ignore that. So basically, there are roughly 85,000 AIS cases annually. 18,000 fusion surgeries is a percentage of 21%. The small difference is probably due to the rapid progression cases that are hard to react to in time, and the unnoticed cases probably due to having a balanced double curve that is hard to identify until it's too late for bracing, and the small number of adult scoliosis cases, etc.

Just a minor point. You can't use the number of new AIS diagnoses each year (assuming ~85k) and divide the number of fusions performed each year (18k) to get a percentage. The number of surgeries should be divided by the total number of AIS cases, e.g. 85k x 9. So 85k/year are added to the total number. Adolescents lasts ~ 10-18 years old. So at any given year there are about 765k AIS cases. From that group, there are 18k surgeries performed. Equals ~2%.

People make the same calculation mistake when they say that half of all marriages end in divorce. If there are 100 marriages every year and 50 divorces, you have to figure out what the average length of time a marriage lasts. This will dictate how big the pool is. Let's say it's 7 years. So in year #1 there are 100 marriages but they haven't gone long enough to get divorced yet. Then year #2 and so on...until year #7. So now there are 700 marriages and now there are 50 divorces. So there's 650 marriages +100 = 750 marriages - 50 divorces the next year = 700+100-50=750 the next year.

Sorry for the random discussion, but that calculation always bothers me.

[Ballet Mom]

Just a minor point. You can't use the number of new AIS diagnoses each year (assuming ~85k) and divide the number of fusions performed each year (18k) to get a percentage. The number of surgeries should be divided by the total number of AIS cases, e.g. 85k x 9. So 85k/year are added to the total number. Adolescents lasts ~ 10-18 years old. So at any given year there are about 765k AIS cases. From that group, there are 18k surgeries performed. Equals ~2%.

People make the same calculation mistake when they say that half of all marriages end in divorce. If there are 100 marriages every year and 50 divorces, you have to figure out what the average length of time a marriage lasts. This will dictate how big the pool is. Let's say it's 7 years. So in year #1 there are 100 marriages but they haven't gone long enough to get divorced yet. Then year #2 and so on...until year #7. So now there are 700 marriages and now there are 50 divorces. So there's 650 marriages +100 = 750 marriages - 50 divorces the next year = 700+100-50=750 the next year.

Sorry for the random discussion, but that calculation always bothers me.

Thanks skevimc...I should have noticed that. I'm glad to hear it's only 2%.

[Pooka1]

Just a minor point. You can't use the number of new AIS diagnoses each year (assuming ~85k) and divide the number of fusions performed each year (18k) to get a percentage. The number of surgeries should be divided by the total number of AIS cases, e.g. 85k x 9. So 85k/year are added to the total number. Adolescents lasts ~ 10-18 years old. So at any given year there are about 765k AIS cases. From that group, there are 18k surgeries performed. Equals ~2%.

So my calculation should have been:

Of the AIS cases, for each year:
30,000 "require" braces.
18,000 are fused

Assuming, kids wear a brace for average of 3 years then that would be 90,000 total kids wearing braces at any point in time.

So would it be 18,000 fusions into 90,000 for brace failures (assuming most of those kids are braced) for a rough estimate of 20%?

If that is correct, that same guy who estimated from that one study that 70% of kids are braced needlessly ALSO nailed the bracing failures!

I would like to meet that guy.

[hdugger]

Of the AIS cases, for each year:
30,000 "require" braces.
18,000 are fused

This calculation assumes that "require braces" = "compliantly bracing"

That assumes 1) that everyone broadly estimated to require bracing is given a brace prescription and 2) that everyone receiving the prescription is compliant throughout the course of the prescription.

Those seem like two very large assumptions to make.

If you want to try an equivalent thought experiment, calculate the number of adults "requiring surgery" and the number of adults with curves above the surgical threshold. We can probably agree that that percentage is not a fair indication of surgical failure.

[Pooka1]

This calculation assumes that "require braces" = "compliantly bracing"

That just makes it equivalent to the study where the 20% failure rate was generated. That is a good thing assuming the compliance rate was about the same. Both numbers are generated from an "observed" bracing + compliance." So the same ratio would result if the compliance rate was low in both cases or high in both cases. It just has to be similar in both cases. It is not a final number on bracing efficacy.

That assumes 1) that everyone broadly estimated to require bracing is given a brace prescription and 2) that everyone receiving the prescription is compliant throughout the course of the prescription.

Those seem like two very large assumptions to make.

There are always plenty of assumptions in this game. This is no exception. The point is that when independent lines of evidence converge, it strengthens both.

If you want to try an equivalent thought experiment, calculate the number of adults "requiring surgery" and the number of adults with curves above the surgical threshold. We can probably agree that that percentage is not a fair indication of surgical failure.

That's not equivalent. Why would it be?

[Ballet Mom]

So my calculation should have been:

Of the AIS cases, for each year:
30,000 "require" braces.
18,000 are fused

Assuming, kids wear a brace for average of 3 years then that would be 90,000 total kids wearing braces at any point in time.

So would it be 18,000 fusions into 90,000 for brace failures (assuming most of those kids are braced) for a rough estimate of 20%?

If that is correct, that same guy who estimated from that one study that 70% of kids are braced needlessly ALSO nailed the bracing failures!

I would like to meet that guy.

I think without the actual numbers we're just playing guessing games. You can't just assume everyone was wearing a brace who gets fused. There are doctors who don't believe in bracing...I suspect their surgical rate is quite high. In the one surgeon's office I visited who didn't like bracing, I met lots of people who were just waiting to grow enough to be fused, without bracing. Some kids and families refuse bracing. Stapling still has the same fusion rate as bracing. Your first daughter wasn't even braced before fusion.

I've also seen a huge variety of times in brace, the standard doesn't even require bracing if you're diagnosed six months post-menarche, which I think is crazy. It's pretty obvious kids curves are increasing even past physical maturity to surgery, let alone six months after menarche. I personally think the US system needs to do a better job of figuring out how and when to brace then just chalking up all the scoliosis surgeries to an obviously flawed system of bracing.

And I still disagree that 70% of kids are braced unnecessarily. That is your opinion.

[hdugger]

I'm likewise suspicious, but I lean towards being suspicious by nature.

I would very much like to hear an evaluation of the test by an impartial SRS surgeon. I know Hart isn't signed up to administer it yet - maybe I'll ask him.

I'd missed the actual quote which stated that Lonner had stock in the scolioscore company:

"Axial Biotech, the company that developed the test, is now working on versions that would apply to other ethnic and racial groups. The current version is based on tests of nearly 10,000 patients at 85 clinical sites around the world and is available at more than 50 spine U.S. centers. (Lonner has stock options in the company.) "

http://www.heraldtribune.com/article...1010?p=3&tc=pg

I generally like entrepreneurs and doctors as separate disciplines, but I really don't like entrepreneurial doctors. Don't we pay these guys enough of a living wage that they could forgo the ethical compromises?

[Pooka1]

According to the authorship line on the validation study published in Spine, Lonner doesn't appear to have analyzed any samples, crunched any data, nor made any conclusions based on same.

ETA: Can someone with the paper check if Lonner is mentioned in the acknowledgment section and, if so, what for?

Thanks in advance.