Why I decided to brace my daughter with the SpineCor

[Pooka1]

(snip)
gotta keep in mind that 10% Utah thing

Titanium Ed lives sort of near Utah (I had to check a map first though). I'm hoping he can help us out with understanding that statistic.

[concerned dad]

I do appreciate your thoughtful comments. I was just kidding about the holding my feet to the coals stuff. It is important to me to convince myself that I am not “drinking the Kool-Aid” when it comes to the SpineCor. There certainly is a risk that a parent like myself, after facing the new news about the diagnosis, will be more susceptible to embracing miracle cures because that is what we want. I don’t think that is the case here. Your comments continually point me in new directions to evaluate and attempt to understand the scientific literature on the subject.

My comments on your last post are in red below.

For the benefit of anyone following this, we are discussing two papers. The first is written by Coillard in 2007 and published in the Journal of Pediatric Orthopedics. Coillard was part of the team that developed the SpineCor. I’m referring to this above and below as “the Coillard paper” and Sharon linked to the full text in a post above.
Effectiveness of the SpineCor Brace Based on the New Standardized Criteria Proposed by the Scoliosis Research Society for Adolescent Idiopathic Scoliosis
The second is written by the SRS Committee on Bracing and Nonoperative Management And published in the Journal Spine in 2005. I’m referring to this above as the “new SRS Guidelines” paper and below as the “new SRS paper”.
Standardization of Criteria for Adolescent Idiopathic Scoliosis Brace Studies
There are other papers that come into this discussion below and I’ll try to keep everything straight.

Now, Sharon, I don’t mean to offend you on this first point. But I would bet my last dollar that you haven’t read (at least recently) the “new SRS paper” in its entirety. If you had, I don’t think you would have made several of your arguments as the SRS paper is pretty clear on many of the issues under debate.

Quote:
Originally Posted by concerned dad
(snip)
In terms of what I think of the design? Well, I think the design was based on the SRS guidelines. Dr. Coillard addressed every point in her paper that was called for under the SRS guidelines.

I think these would be correctly referred to as criteria for comparing different braces

Criteria or Guidelines? I think it is semantics. But the “new SRS paper” says (emphasis added is mine)

The purposes of our study are to define consistent parameters for inclusion criteria for orthotic treatment of patients with AIS and to define consistent parameters to assess the effectiveness (outcomes) of treatment. These parameters could then serve as guidelines for all future AIS bracing studies to make comparisons among studies more valid and reliable.

OK, minor point but I will stop referring to it as I did above as the new SRS Guidelines paper and just say the “new SRS paper”.

What I think they are assuming is that controlled studies are needed to say anything about bracing efficacy over and above watching and waiting.

No, this is not at all what they are saying in that paper.


Why do I think the SRS assumes a controlled study is necessary for this purpose? Because they say so...

In 1985, the Scoliosis Research Society (SRS) initiated a study to investigate the effectiveness of bracing as a treatment for scoliosis. Many previous studies of full time bracing showed that braces stop about 80% of curves. All of these studies, however, were "uncontrolled" which means there were no simultaneous groups of untreated, unbraced patients for comparison. Therefore, there was some doubt that brace treatment of scoliosis was effective, and concern that bracing may be no different than "natural history" or what happens when no treatment is undertaken.

Yep, the SRS said that in 1985, but it is not at all what they said in 2005 in their “new SRS paper”. But the quote you printed above brings up a really relevant point. What happened after they “initiated a study” ? They indeed did complete a study. Even better, a controlled study. It was presented by Nachemson in 1995 (10 years later in the Journal of Bone and Joint Surgery). What did the study conclude? “Treatment with a brace was successful”. Why is it relevant? I need to provide some background.

In an earlier thread (linked in my first post) Sharon led me down the path of enlightenment concerning the debate about bracing efficacy. I had originally assumed that it was a settled matter but that is not the case. My “take” on this is that Dolan and Weinstein at the University of Iowa made a case in an article published in Spine in 2007 that, based on their review of the literature, the efficacy of bracing is undetermined. But this is the thing, I am completely blown away that they excluded the, 10 year in the making, SRS supported, Nachemson study which contained 129 unbraced patients. Rather, they used data from just 2 centers that totaled 30 unbraced patients to come to their conclusion. Talk about the potential for cherry picking data. What was the outcome of their research? Guess who the principal investigators are for the current BrAIST study. Now there may be some very valid reason for all this, but before I would agree to be part of that random study I would want to have much more clarification on the issue of why the supporting study was designed in such a way as to exclude the Nachemson data. Maybe we can talk about the Nachemson paper in more detail later. EDIT after further review, I see that I am wrong. Continue reading to see this discussed further in the thread.
But I digress……

I repeat my concern... I don't see how you know what the bracing is doing without a control group.

We have beaten this issue to death. Retrospective studies do not have control groups per se. I agree that is a weakness. The “new SRS paper” attempts to strengthen future retrospective studies by establishing guidelines/criteria. Your argument is nearly identical to the arguments tobacco companies used to say there was no proof cigarettes caused cancer. I maintain that the reason there are so few controlled studies relates to ethics. The 2007 Iowa paper was essentially used to negate the ethical concerns and pave the way for the current random controlled study.

In this paper (we’re back to Coillard now), they are claiming victory mainly with smaller curves that are known not to progress.

No, Coillard excluded 112 of their patients with a Cobb angle less than 25 degrees from their analysis. Why did they do this? Because the “new SRS paper” (I would really prefer to refer to it as the SRS Guideline paper) said that only patients with initial curves between 25 and 40 degrees should be included in the study. This is detailed on the first paragraph of the second page of the Coillard paper.

But because the patients were braced they mark it in the "successful bracing" column. For all we know, none of those patients with smaller curves or even some with larger curves would have progressed anyway. Nothing in this study rules out that possibility. I continue to question the peer-review and editing of these journals.

No, not even one of the patients with an initial curve less than 25 degrees was included in their analysis.

Quote:
Originally Posted by concerned dad

“Comparing the end of bracing Cobb angle to the one at 2 years after bracing, our study reveals that the follow-up of orthopaedic treatment was a success in 95.7% of the patients, with a mean correction angle of 8.6 +/- 1.7 degrees.“.

No actually I think that statistic is quite unclear and you were right to be confused. I am still confused as to what exactly it means.

I am pretty sure I have it right. It reflects the percentage of patients initially considered to HAVE BEEN successfully braced and still had to go on to surgery within 2 years. A parameter specifically called for by the “new SRS paper”.

Here are some questions I have about that statistic:

1. It's important to realize that only 47 of the 170 patients (~28%) in the study were followed up at 2 years. Why so few? Were these predominantly the ones who started with smaller curves that wouldn't have progressed anyway?

Well, we know (as I outlined above) that they were not the ones with the smaller curves. They are the ones who both hit the 2 year post bracing mark and achieved a “successful outcome”. So, it certainly excludes the 39 patients who had surgery and it excludes the 12 who withdrew from the study. It is unclear if it excludes the 14 who were weaned before reaching skeletal maturity (although I would bet that it does.) So that still leaves 105 patients who are either still within the 2 year window or who didn’t travel back to Montreal to be monitored after bracing.

This got me thinking about updates. I went on pubmed to see if anything new was out there and found a new 2008 paper by Coillard. Instead of 47, they now have 162 patients who have reached that point. I haven’t fully waded through that paper yet.

continued on next post

[concerned dad]

continued from above

2. I don't want to be incendiary but this is an overt case of data selection EVEN IF only 47 patients were available to them. Ask yourself what is the lower limit of percent of total patients where it wouldn't matter how many were stable?

That’s not incendiary (I was originally incendiary above in my criticism of Dolans paper – I toned it down a bit). Yeah, 47 is not a large sample, but they now have 162. That’s all they had at the time. And, I have to say, I’m glad they published their results instead of waiting. Sure, they could have presented a stronger case with more data but if you have something you believe will help children, maybe they felt they had a moral obligation to get the current results out there. I ask again, where are the bracing studies for other braces (reported in accordance with the new SRS guidelines so we parents can make actual comparisons)?

For me, ~28% is too low to draw any conclusions given that we know smaller curves will not progress despite no treatment and that we know several treated kids went on to have fusion.

Smaller curves were excluded (sorry to drive this point home). We know that, no matter what, some curves will progress.

3. What exactly do they mean by weaning point? Is that at the start of the weaning or at the end? Elsewhere they say the comparison is at the end of bracing. I take that to mean before the weaning period. How long is the weaning period? Is it 2 years or some substantial portion of that time? Why is a weaning time necessary if the brace is working? I think what needs to be compared CLEARLY is the end of the main bracing (allowing a week solid out of brace) and two years after that point (assuming the weaning period isn't more than a few months).

Good question. I don’t understand the weaning thing. Maybe it is discussed in the bracing manual on the SRS website. But, I would venture a guess that it is NOT a substantial portion of the 2 year post bracing time. The fact that they show no/little change over those 2 years post bracing seems to make the matter moot.

The two year period that the Montgomery paper talks about almost certainly doesn't involve any weaning period which is really a continuation of bracing.

You are correct, if indeed it takes two years to wean off the brace. I’m going to see if I can find out more about this weaning stuff. I see in the new paper I just downloaded, they are also reporting results 5 years post bracing (in addition to the “SRS called for” 2 year data).

Other questions I had about this paper include:

1. no control group (as mentioned above)

No further comment

2. extremely liberal indication for surgery of >60*! Why not 80*? Or 100* This is ridiculous as most surgeons pull the trigger at far lower Cobb angles. How would have the stats changed if they adopted a more realistic point like 45* or 50*. We scheduled surgery when my daughter was in the high 40s* (albeit with a known high rate of curvature).

To account for this, The “new SRS paper” calls for a reporting of how many patients didn’t meet the “indication for surgery” yet still progressed beyond 45 degrees. Coillard reported it, and it is 2 patients.

This is further proof that surgery recommendation or having surgery is an extremely non-robust (i.e., worthless) criteria.

You have to admit it is less “worthless” if they also tell you how many fall in between. Different countries have different criteria for surgery.

3. 101 of the 170 patients were reported to be stable after bracing. But when was that determined? Was it determined within 5 minutes of taking the brace off for the last time? Was stability determined before, during, or after the weaning period?

Good questions. You would make a good journal reviewer. It’s too bad none of the official reviewers picked up on this and pressed for clarification. But, so long as the weaning period isn’t 2 years, does it really matter? I mean, come on, the outcome for the successfully treated patients at 2 years is dramatic.

4. It is important to emphasize that these results at best strictly only apply for the group of patients that meet all criteria. 144 kids being treated didn't meet the criteria. It would be interesting to know anecdotally how they failed to meet

That’s an easy one. The criteria they failed to meet were, of course, the criteria in the “new SRS paper”. Those 144 kids were either: younger than 10, had a Risser of 3 or above or were more than 1 yr post menarchal, had a Cobb angle less than 25 degrees or more than 40 degrees. But I agree, the results only strictly apply to that specific group. However, that doesn’t mean that the results can not apply to those outside the group.

and generally how this group is doing.

Christine2 has a daughter that didn’t fall within the SRS guidelines for reporting and wasn’t included in this study (or the brand new one I just downloaded). Her curve went from over 30 degrees to about 1 degree right now. Anecdotal? Yes. But I’ll take some of what she’s having for my little girl.

[christine2]

KUDOS Concerned Dad

I wish everyone on this forum could have the great results my daughter is having. We just have to be diligent to keep it that way and pray that we can make it through the growth spurt.

[Pooka1]

I do appreciate your thoughtful comments. I was just kidding about the holding my feet to the coals stuff. It is important to me to convince myself that I am not “drinking the Kool-Aid” when it comes to the SpineCor. There certainly is a risk that a parent like myself, after facing the new news about the diagnosis, will be more susceptible to embracing miracle cures because that is what we want. I don’t think that is the case here. Your comments continually point me in new directions to evaluate and attempt to understand the scientific literature on the subject.

It's important to recall that this isn't my field. I'm making what I consider general, universal design comments. I have questions about how this stuff is done and interpreted and I may be missing some understanding.

Now, Sharon, I don’t mean to offend you on this first point. But I would bet my last dollar that you haven’t read (at least recently) the “new SRS paper” in its entirety. If you had, I don’t think you would have made several of your arguments as the SRS paper is pretty clear on many of the issues under debate.

Things like that are not inherently offensive. And you are correct. I can't find the full text of that paper. I have read the abstract. I don't see anything there as a departure from their earlier criticism of uncontrolled studies. I see it as an extension to further make progress in the area of meaningful study design. Does the full text say anything about controlled studies? Recall the site I posted about various study designs and which are viewed as better and best.

That said, even if they came out and said they were abandoning the call for controlled studies (which I doubt they did or will), I still question it. How do you know any of the bracing is working over and above watching and waiting without a control group? Anyone feel free to answer that at any time.

Yep, the SRS said that in 1985, but it is not at all what they said in 2005 in their “new SRS paper”.

No I don't see the 2005 criteria for brace comparison studies as supplanting the earlier call for controlled studies. There is no getting around the need for controls. If you think there is, please explain.

But the quote you printed above brings up a really relevant point. What happened after they “initiated a study” ? They indeed did complete a study. Even better, a controlled study. It was presented by Nachemson in 1995 (10 years later in the Journal of Bone and Joint Surgery). What did the study conclude? “Treatment with a brace was successful”. Why is it relevant? I need to provide some background.

Coincidentally, I posted this paper a little while ago for discussion. This paper appears to have serious issues in my opinion.

In an earlier thread (linked in my first post) Sharon led me down the path of enlightenment concerning the debate about bracing efficacy. I had originally assumed that it was a settled matter but that is not the case. My “take” on this is that Dolan and Weinstein at the University of Iowa made a case in an article published in Spine in 2007 that, based on their review of the literature, the efficacy of bracing is undetermined. But this is the thing, I am completely blown away that they excluded the, 10 year in the making, SRS supported, Nachemson study which contained 129 unbraced patients. Rather, they used data from just 2 centers that totaled 30 unbraced patients to come to their conclusion. Talk about the potential for cherry picking data. What was the outcome of their research? Guess who the principal investigators are for the current BrAIST study. Now there may be some very valid reason for all this, but before I would agree to be part of that random study I would want to have much more clarification on the issue of why the supporting study was designed in such a way as to exclude the Nachemson data. Maybe we can talk about the Nachemson paper in more detail later.

I think there is very good reason for not including the Nachemson and Peterson paper. They were correct to exclude it.

Retrospective studies do not have control groups per se.

I don't see why not. If you can assemble a treatment groups retrospectively, why can't you assemble a control group in the same manner??? This is outside my field and there may be some esoteric reason but I can't imagine what it is.

I agree that is a weakness.

That's an understatement. How do you know if the brace is doing anything without a control group? It could be that the brace is doing nothing but you wouldn't know it with out a control group.

The “new SRS paper” attempts to strengthen future retrospective studies by establishing guidelines/criteria. Your argument is nearly identical to the arguments tobacco companies used to say there was no proof cigarettes caused cancer. I maintain that the reason there are so few controlled studies relates to ethics. The 2007 Iowa paper was essentially used to negate the ethical concerns and pave the way for the current random controlled study.

I'm sorry but I am not following here. It is simplicity itself to compare (huge) populations of smokers ("treatment" group) and non-smokers (control group) even today.

In re "smaller" curves...

No, Coillard excluded 112 of their patients with a Cobb angle less than 25 degrees from their analysis. No, not even one of the patients with an initial curve less than 25 degrees was included in their analysis.

By smaller I mean at the low end of the range they studied. There is quite a large jump in the likelihood of curve progression in curves below 30 or 35 let's say and those above 40*. It's an exponential (or nearly so or power function or something) curve. Thus small increments in curves translate to relatively larger increases in the chance of progression. Therefore I'm guessing many/most of the "successfully braced" curves started out at the beginning of the study at the low end of the studied range.

I have no comments on your second post.

[Pooka1]

Spontaneous complete correction

Christine2 has a daughter that didn’t fall within the SRS guidelines for reporting and wasn’t included in this study (or the brand new one I just downloaded). Her curve went from over 30 degrees to about 1 degree right now. Anecdotal? Yes. But I’ll take some of what she’s having for my little girl.

I stumbled across this when looking for some progression curves...

Heary and Albert, p.95 (bottom right)

A 5-year prospective study of 85,622 kids found 1,436 kids with scoliosis.

Results over an average observation of 3.2 years:

Progression - 14.7%
Spontaneous improvement of at least 5* - 27.4%
Complete spontaneous resolution - 9.5%

Here is yet another reason why controls are needed. About twice as many kids improved compared to the kids who got worse. And significantly, almost 10% of the kids in this study had a complete spontaneous resolution of the scoliosis.

Had they happened to be enrolled in a Spinecor study at the time, these spontaneous improvement and resolutions would have be chalked up to the efficacy of the Spinecor brace. The bracing literature that doesn't address controls is clearly part of the reason why most published research results are false.

We can't assume that any child in a Spinecor brace would not have completely resolved on their own absent the brace.

This literature has to be cleaned up quickly in my opinion.

[leahdragonfly]

brace non-compliance ignored in studies

Hi Concerned Dad,

I haven't had time to read all the posts here thoroughly, so pardon me if this topic has been discussed. Clearly the bracing literature is very controversial. As parents we each have to decide what treatment we feel is best for our child at the time. For some of us it is extremely important to feel our decisions are backed by sound scientific research--others, less so.

One major problem I see with bracing research is that the issue of non-compliance with the prescribed brace wear is almost completely ignored. All children who were in the braced groups were ASSUMED to have been 100% compliant with their brace-wearing. Since 23/7 bracing (with a hard brace anyway) is almost impossible to consistently achieve in the real world, it is a very dangerous assumption to make. It means that some kids' curves that are chalked up to being successfully braced, stayed stable even though the child did not wear the brace as much as prescribed. I was one of those kids. To me this problem alone invalidates most of the bracing research, as it gives false credit to bracing as a successful treatment when in fact the brace wasn't the reason the curve stable. Does this make sense?

At least the BrAIST study that is underway seems to address this by having temperature sensors in each brace to accurately determine brace wear.

[Pooka1]

Excellent point

I agree with Gayle that compliance is a major confounder in the brace literature.

But I see the result being a potentially false negative and that braces might in fact be more effective than we can presently show.

That is, if everyone wore the brace the prescribed number of hours, it certainly might be the case that bracing is more effective than watching and waiting. But we can't know unless we get a handle on the compliance within a controlled study.

I have been thinking about this and I think the use of averages when there is so much natural and other variation is not going to be fruitful. I think it's time to go to individual-based models. The issue I brought up early about the propensity of curve increase related to degree of curve makes this seem obvious. When you add that Risser is also known to be highly correlated with propensity of curve progression, then I think that, with all the other reasons, calls for individual-based models.

Now I am not a statistician and I have never used these types of models. But I have taken note of the fact that certain folks in fields related to mine have gone to these models for a reason. As far as I can tell, bracing study results would be far more robust if using these models. Averaging over smallish populations with several known confounders is not likely going to produce clear, "true" results as far as I can tell.

[concerned dad]

I stumbled across this when looking for some progression curves...

Heary and Albert, p.95 (bottom right)

A 5-year prospective study of 85,622 kids found 1,436 kids with scoliosis.

Results over an average observation of 3.2 years:

Progression - 14.7%
Spontaneous improvement of at least 5* - 27.4%
Complete spontaneous resolution - 9.5%

Here is yet another reason why controls are needed. About twice as many kids improved compared to the kids who got worse. And significantly, almost 10% of the kids in this study had a complete spontaneous resolution of the scoliosis.

That was an interesting book you linked to. Too bad google doesnt offer the complete text, but there is a lot there still.
Regarding that specific study you cited, I looked up the abstract (couldnt get the entire paper). But, it looks like they included kids who had curves as low as 10 degrees.
Curve progression was studied in 839 of the 1,436 children with idiopathic scoliosis of at least 10 degrees detected from the school screening program
We know many small curves dont progress. Again, this is part of the reason for the SRS guidelines calling for studying bracing efficacy in curves between 25 and 40.
They didnt stratify their results in the abstract, perhaps in the paper, but I would GUESS that the majority of spontaneous corrections occurred in the smaller curves.
they do say
More specifically, the following were associated with a high risk of curve progression: <snip> and curve magnitude (> or = 30 degrees).

Had they happened to be enrolled in a Spinecor study at the time, these spontaneous improvement and resolutions would have be chalked up to the efficacy of the Spinecor brace. The bracing literature that doesn't address controls is clearly part of the reason why most published research results are false.

We can't assume that any child in a Spinecor brace would not have completely resolved on their own absent the brace.

All those kids would not have been enrolled in the study, the SRS guidelines would have excluded the vast majority

This literature has to be cleaned up quickly in my opinion.

The SRS would seem to agree with you, that's why they established the guidelines for future bracing studies. To make them MORE meaningful. Not to make them 100% conclusive (as a random controlled study would approach), but to improve the interpretive significance of the literature.

The SRS guidelines basically said, (liberally paraphrasing here)
hey folks. listen up. if you're going to continue publishing these retrospective uncontrolled studies, at least use consistent criteria so we can try to make some sense out of all this.

and so far, 4 years later, the ONLY people who are willing to present their data in accordance with the new guidelines are the SpineCor folks.
And there has to be a ton of data for the other braces. Where is it?

I do have some comments on your other post, and I do want to discuss Nachemson. Right now I'm working my way through the new Coillard paper that updates their results since 2007.
Might I suggest we keep the discussion to this thread. We are probably boring the heck out of people.

[concerned dad]

I agree with Gayle that compliance is a major confounder in the brace literature.

But I see the result being a potentially false negative and that braces might in fact be more effective than we can presently show.

Thank you. Yes, it is a confounder. It could affect results one way or the other.

[concerned dad]

I have been thinking about this and I think the use of averages when there is so much natural and other variation is not going to be fruitful. I think it's time to go to individual-based models. The issue I brought up early about the propensity of curve increase related to degree of curve makes this seem obvious. When you add that Risser is also known to be highly correlated with propensity of curve progression, then I think that, with all the other reasons, calls for individual-based models.

Wow, Sharon, I'm impressed. Even though you didnt read the full SRS guidelines, you could have been the one to write them.

This is exactly why they call for (actually, I think they use the word 'encourage") reporting the data using stratification of results.

[Pooka1]

That was an interesting book you linked to. Too bad google doesnt offer the complete text, but there is a lot there still.
Regarding that specific study you cited, I looked up the abstract (couldnt get the entire paper). But, it looks like they included kids who had curves as low as 10 degrees.

Yes. That was implied when they said these kids had scoliosis. That is the minimum angle everyone uses for the definition as far as I know.

(snip)

All those kids would not have been enrolled in the study, the SRS guidelines would have excluded the vast majority

Maybe but we don't know that. We need to see the distribution of curves and the distribution of curves that completely straightened and follow individuals perhaps. This (brace research in general) may be an obvious instance where individual-based models are needed. I think it is but I don't work with this kind of stuff so I don't really know.

(snip)
The SRS guidelines basically said, (liberally paraphrasing here)
hey folks. listen up. if you're going to continue publishing these retrospective uncontrolled studies, at least use consistent criteria so we can try to make some sense out of all this.

and so far, 4 years later, the ONLY people who are willing to present their data in accordance with the new guidelines are the SpineCor folks.
And there has to be a ton of data for the other braces. Where is it?

Perhaps they have decided to stop publishing uncontrolled studies? Can you tell me how you know anything of what a brace is doing is you don't have a control group?

I do have some comments on your other post, and I do want to discuss Nachemson. Right now I'm working my way through the new Coillard paper that updates their results since 2007.

Can you post this paper?

[Pooka1]

Wow, Sharon, I'm impressed. Even though you didnt read the full SRS guidelines, you could have been the one to write them.

This is exactly why they call for (actually, I think they use the word 'encourage") reporting the data using stratification of results.

I'm not sure but I don't think stratifying the results is the same as individual-based modeling but it seems like it as a step towards that. We need someone who knows about this stuff to comment.

[concerned dad]

Can you post this paper?

I am trying to find a publicly available link. So far, no luck. I bought the paper for 15 euro's from the publisher. Looks like a paper from a sosort conference. I imagine that soon it will be available on the Spinecorporation website.
I could attach it here but am afraid it would violate copywrite laws and NSF would need to remove it.

But, while looking, I came acrosss something disturbing/confusing.
here they have reference to weaning. Look at page 37. They have a an xray of a patient there with text "weaning time:62 months"

Now to me, weaning means a gradual withdrawal of something. I wonder if this is a language thing or if indeed this patient has worn a brace for a large part of those 62 months following the "weaning date".

Are there any forum participants who have been "weaned" off the SpineCor who could shed some light on this? I am thinking it is a language thing, but as Sharon points out above, IF the brace was worn after treatment ended then there are some serious implications for assessing the "permanent" aspect of success

[concerned dad]

OK, it looks like the weaning period takes 6 months.
At the weaning evaluation visit, the patient is xrayed (after not wearing the brace for 72 hours).
Then, the brace is put on and another in brace xray is taken
if the two differ by less than 5 degrees, weaning commences

Weaning consists of wearing the brace 10 hours a day for 6 months, then
show up for appt after NOT wearing the brace for 3 days (72 hours).
Take out of brace xray

Then take in brace xray

If difference is less than 5 degrees, complete discontinuation of brace.