Wii Balance Board Gives $18,000 Medical Device a Run For Its Money 422
Gizmodo highlights a very cool repurposing effort for the Wii's Balance Board accessory. Rather than the specialized force platforms used to quantify patients' ability to balance after a trauma like stroke, doctors at the University of Melbourne thought that a Balance Board might serve as well. Says the article: "When doctors disassembled the board, they found the accelerometers and strain gauges to be of 'excellent' quality. 'I was shocked given the price: it was an extremely impressive strain gauge set-up.'" Games controllers you'd expect to be durable and at least fairly accurate; what's surprising is just how much comparable, purpose-built devices cost. In this case, the Balance Board (just under $100) was compared favorably with a test platform that costs just a shade less than $18,000.
Re:"Not for ________ use" (Score:2, Informative)
Another question to ask is exactly how much time/effort/money Nintendo went through to get this controller FDA approved. What? It's not FDA approved?
Re:"Not for ________ use" (Score:5, Informative)
exactly. the FDA requires significant documentation of the hardware and software along all lines of the R&D, and manufacturing process. which are actively audited by the FDA. documentation, and documentation compliance is a huge chunk (not the largest, but definitely a line item on their accounting paperwork) of their budget.
Re:What have we here? (Score:3, Informative)
As a guy who designs electronics for load transducers for living, I can tell ya that a balance board equivalent platform won't run you anywhere near $18,000. Probably more like $4k. One thing that makes the Wii device cheap is the mass production. Try selling a 100 of those a year for $100 apiece -- it's impossible. The tooling for plastic injection molding will cost you more than $18k alone. As for electronics themselves, they are not really a factor in the supposed $18k, er, $4k, price. I admit I haven't checked how accurate the balance board is. All I know is that $4k buys you a platform that has ~0.5mm accuracy for center of pressure across its surface of roughly 1/4 m^2. The resolution is better than that, of course.
OTOH, I'm going out and buying the darn balance board right about now :)
Re:"Not for ________ use" (Score:3, Informative)
It used to be the same with "unix" hardware in the 1980's/1990's, particularly with commodity components like mice, printer and scanner cables. An RS232 cable for an Apple Mac or PC clone that cost around $20, would be marked up for $200 for a UNIX workstation. To make sure that only the official cable was used, there would be loop-back configurations built into unused pins at each end of the cable, so that a connector patched up from twenty-five core cable and a couple of RS232 snap connectors wouldn't work.
accelerometers? (Score:3, Informative)
"When doctors disassembled the board, they found the accelerometers..."
They did? I couldn't find any information stating that the balance board had motion sensing. Everything I've read says it just has four pressure sensors, one for each corner and that's it.
Re:accelerometers? (Score:4, Informative)
New Scientist may have made that up. The actual journal article on ScienceDirect only discusses a black-box approach: i.e., putting patients on the board and seeing what happens. The procedure doesn't mention disassembling the device whatsoever!
Re:"Not for ________ use" (Score:3, Informative)
I call BS. I have been involved in medical device development and licensing for some time. I can ballpark what the liability for a device such as this would be, and it is pretty small. The reality of the medical device industry is that it is a bubble that is largely propped up by low expectations, high regulatory barrier to entry, and other inherent consequences of our current medical device market system.
Take, for example, a glucose monitor (specifically, the most widely used system on the market in the US). The consequences of a technical failure are relatively high, including severe injury to the user. However, the insurance coverage for this type of device is barely 5% of the total retail cost of the sensor unit, and just over 8% of the retail cost of the disposable strips. The majority of the cost is from development overhead and ongoing quality assurance. The margins are still huge- fine. Here is the problem: independent testing results bear little in common with the reports filed with the FDA regarding accuracy and precision of the units. In fact, using their own test strips with calibrated samples of human plasma, the units only had a 60% confidence interval of indicating blood sugar to within within 15 points and the precision was not believable to within 20 points. Subsequent anecdotal testing by coworkers was consistent with our lab results (we were making an Iphone wart to to perform glucose management well before it seemed everyone wanted to do the same thing). One person saw blood sugar go from 70 to 110 to 87 within 10 minutes just because he tested with three different monitors with the same lot of strips. Our monitor was within a few points over 95% of the time, with about 3-point tall error bars.
The rub in this is that our device was going to sell for $12 versus $75, give superior results, and link to a wealth of online management tools. We couldn't do it because of the regulatory barriers, and even more because insurance companies would be slow to adopt a new device.
There are other examples in every doctor office. Hand/eye coordination test software that has less sophistication than PONG, or the first interactive app your kids ever wrote, yet the software costs $4200 plus annual licensing fees and docs bill it at $220 per use. This one is a doozy because I know the guy who wrote it and HE fully acknowledges this from the left seat of his sports car and life that has no other discernible source of income.
While I cannot offer a solution (that would be palatable to the majority of Americans with a shriveled connection to the inner and outer workings of their own lives), this example is not a new thing, but is is an example of something I have a great deal more respect for: innovation linked to something we quaintly used to call value.
As for the actual application, I cannot fathom that the same functionality could not be achieved for far less.
Re:"Not for ________ use" (Score:5, Informative)
Nice post with a nice result, although you used some mystery maths to get there.
Sales proceeds need to be: $93.5 million. That means, the price for each unit needs to be: $93.50 per unit. What if they want a healthier profit margin? Their sole purpose in life is to manufacture medical devices, they don't sell software -- they need a good profit margin from selling their product. A fair profit margin is 100% or more. To achieve that, the minimum price is $930.50 per unit.
First you say 85 bucks is their break-even point, which means they have to sell it at 93.50 for a 10% profit margin (1.1 times 85). Then you somehow (almost) multiplied by 10 to get 930.50 bucks?
If 85 bucks is break-even, then they'll have a 100% profit margin selling at twice that, or 170 bucks. Not 930.50.
Also, 1 million sales is unrealistic for a niche product, it will probably be more like 200,000 sales. To maintain a healthy product with 1/20 of 1 million, the actual price needed will be 20x that, or $18,610.00
200,000 is 1/5 of 1 million. So following your line of thought, the actual price would be 5 times 170, or 850 bucks.
You did, however, say they produced 1 million of them, costing them 75 million in parts and labor. If they're only making 200,000 those costs would only be 15 million. Adding 10 million research costs to that makes a total of 25 million in production costs. To get that 100% profit margin, they'd need to earn 50 million in sales, meaning 50 million bucks/ 200,000 units = 250 bucks per unit.
Holy smokes, that's nowhere near the "$18,000 medical device" price.
Re:"Not for ________ use" (Score:5, Informative)
You have obviously never worked in an industry with high quality control standards.
I worked at a nuclear power plant that happened to be on a Navy ship. One of our supply guys had to order some simple part (a resistor or something) that was categorized as part of the nuclear plant, and it was maybe $100 through the supply system. He took the part number home, called the supplier from his home phone presenting himself as a hobbyist. It was about $1 for the same part. I guess we were paying $99 for the QC sticker they put on it before they sent it off through the supply system.
Disclaimer (because /. has taught me I need one): I'm not saying all QCed stuff is expensive for no reason. Just that sometimes people really are getting ripped off in the name of QC.
Comment removed (Score:5, Informative)
Re:"Not for ________ use" (Score:1, Informative)
"Sterile out of the box" is such a stupid fucking point I can't believe you typed it without your brain dribbling out of your nose and short-circuiting your keyboard. If a guy with an open foot ulcer catches an infection from standing on the scales, it's the doctor's fault for a) not sterilising the scales b) not ensuring the open foot ulcer is bandaged like it ought to be. Hey, if a scalpel packer sneezes on a scalpel being packed, then the surgeon takes the scalpel and sticks it in somebody's liver without cleaning it, who's at fault? The scalpel packer according to your retard logic!
Most of your other points are pretty decent, I just can't believe you decided to lead with such a non sequitur.
Re:sounds familiar (Score:2, Informative)
Re:"Not for ________ use" (Score:1, Informative)
So why? I don't get it.
Because Americans don't understand how insurance companies work. Insurance agencies broker risk, and they do that most cheaply with an economy of scale: the larger the pool of risk, the more predictable (cheaper to manage) it becomes. But the insurance market is (rightfully) regulated. An important example is a capitalization requirement, which limits the number of clients an insurance company can take on. In order to "get around" this requirement, insurance companies by and sell risk with other companies, through the use of "swaps". This is perfectly legal, and is in fact a VERY smart move.
But there are two negative consequences for the rest of us: using swaps to settle debts -- instead of itemized bills -- incentivizes fraud, which puts strong upwards pressure on prices. The insurance companies don't care that they're being overcharged, because they're overcharging each other by similarly inflated prices. Secondly, the use of swaps to settle debts is a de facto price fixing mechanism. Honestly, I see this as a benign consequence. The insurance companies are all getting similar actuarial data, and it is based on this actuarial data that they set their prices and negotiate swap contracts. So, in the end, a national-scale insurance company or cooperative would be compiling this same data, and fixing their prices in the same way.
Competition does not enter into the insurance industry, with mathematical force.
Re:"Not for ________ use" (Score:2, Informative)
I did what you're doing for a living awhile back.
The 510K is an absolute joke. What you're doing when you claim a 510K exemption, is you're claiming your new medical device is 'substantially equivalent' to a device already approved. Saves you the astronomical cost of Clinical Trials. So you have your Regulatory Staff out working on that. Meanwhile your Marketing Staff is working double time to convince the customers that your produce is a new, exciting, innovative product.
Re:"Not for ________ use" (Score:2, Informative)
Re:"Not for ________ use" (Score:3, Informative)
It's not the manufacturer's liability insurance that causes those high prices, it's the patients' medical insurance. The only reason they can set the price that high is because it just gets billed to the insurance. Then they inflate the MSRP even more so that they can still bilk the insurance companies even after a "discount."
When my mom was shopping for hearing aids, a particularly honest sales droid told her that the "MSRP" (which no one ever pays) was $12k for a certain model. They discounted it to $8k when they billed insurance companies, but if you didn't have insurance, they'd sell it to you for $3k. So basically they get an extra $5k profit if you happen to have insurance.