This is a follow up to my proposal for widespread monitoring and analysis of biomarker data in people to improve medical research and treatment.
There would be work to do to bring costs down and to make more comprehensive tests. Which is why I was suggesting that someone like an Andy Grove (who lead Intel in bringing the costs of semiconductors lower) would be an ideal person to bringing this about. Andy Grove has lamented the lack of pharma progress. He is motivated.
The global market for microfluidic technologies was worth an estimated $2.9 billion in 2005. This figure should grow to $3.2 billion in 2006 and $6.2 billion by 2011. Self monitoring of blood glucose in diabetics is $1.2 billion business.
51 tests for monitoring glucose for diabetics $100. Results downloadable to PC. Have to make more detailed blood analysis and then transmit results for centralized processing.
There is work on lung cancer blood tests.
It is a goal (widespread biomarker tests) to work towards that would provide a lot of benefits. Plus there are earlier stages where it is only for a few thousand people like Nielson boxes. A statistically diverse group that would help researchers to make better inferences about larger populations. Instant clinical trial data from recorded data mine-able information. Information to improve health and lower costs.
I think the initial few thousand Nielson box goal is achievable in a few years and the larger vision within ten years. Lab on a chip are lithography and MEMS. It is a matter of getting HMOs and PPOs and government funded medical programs to see that this would help them lower costs in the long run.
In terms of the costs. The system would be a lot cheaper than doctors or nurses taking the blood samples and sending them to labs for the tests. A really capable machine would probably initially cost several thousands to tens of thousand of dollars. It would have several chips and perform multiple tests and would have re-usability (instead of throw-away systems).
Some have the attitude: "if this was cheap enough then it would already be widely utilized" is the wrong way to look at it. I think this would be a superior way to figure out what is really happening with people when they get treatment and in between exams.
The current presidential candidate suggestions for healthcare which all amount to put more money into the current broken down health care system (110 billion/year for the Clinton plan) and get some more insurance for some or all of the uninsured. These plans do not try to prevent people from getting sick in the first place, or have a central goal of early detection of disease (when it is cheaper to treat and treatment is more successful) or take steps to get the data we need to make people healthier.
Are we trying to change the future for the better or aren’t we ?
$5 billion per year would pay for tens of thousands of machines and a research program for further development and refinement. 5% of the Clinton plan.
Many of the medical tests are priced so that based on the number of tests performed the research costs of developing the test are recovered with profit. A high degree of automation and high test volumes would allow for negotiation to the Intellectual property holders to reduce the per test prices while still allowing for sufficient return on the research.