It's discussed in a journal article from earlier this year by a pathologist:
> panel profiling, which was introduced in the 1970s as a way of identifying early biochemical changes of disease in asymptomatic individuals, had been abandoned in the 1980s, not so much for the cost. It has long been realized that with multiparametric testing, approximately 5% of results will be false positives, i.e., test results outside the reference intervals, in otherwise normal subjects. This is due to the definition of reference intervals, as being values between the 2.5 and 97.5 percentile of a reference (normal) population. The high cost of investigating seemingly abnormal results in normal people, and the added anxiety of patients, has led to the complete replacement of such biochemical profiling with what is now known as “discrete testing”. In the latter, tests are performed by the testing laboratory, only if requested specifically by the physician.
http://www.degruyter.com/view/j/cclm.2015.53.issue-7/cclm-20...
If a lab result is outside the normal range and a patient has a reason for it to be outside the normal range, that's fine. If a patient has a deviation that is related to their presenting complaint, that's fine too. If a patient has an unexplained deviation which is not related to the presenting complaint, well, there's a 1 in 20 chance that the variation is statistical only. So you use your clinical judgement. is this a patient who requires outpatient follow-up and repeat, or is this an abberation?
Most people simply aren't smart or informed enough to understand things like "false positives".
This would be what I see as the fundamental flaw behind Theranos, even if their technology worked (works?) as promised: the idea that more data will always be better is sophomoric (and I select that word judiciously, considering the founder). Evidence-based medicine clearly shows that more data does not, in itself, lead to a better outcome.
Studies [1] show that weekly INR testing results in a 55% reduction in strokes, 35% reduction in major bleeds and 39% reduction in deaths.
[1] Henagen C. et al. Lancet 2006; 367:404-11
Smart people apparently claim that blood pressure is one of the most reliable indicators of how long you'll live. If so, then it's always seemed strange it's (almost) never measured.
It's a very consumer-targeted technology, although it would certainly find a place in emergency departments and ICUs. I am at a loss to think of ways as to how we would actually capture the data, although better minds than mine I hope come up with ways.
The problem is that to get a good read on arterial pressure you either need to do it the old fashioned way (occlude the artery and record that pressure, then slowly drop it until it's constantly flowing again - see [0]) - or you need to stick a cannula into an artery, as we do in ICUs, and measure pressure using a transducer.
Even technologies that stress their 'passivity' (see [1]) and try to capture this market use the old fashioned way. I don't see that changing anytime soon - you could try and somehow monitor the stretch of a small artery maybe using some variation of current o2 saturation sensors, coupled with advanced computer models of flow rate and variation in small arterioles, but that is a world away and would seem to me to be highly subject to variation/sensitivity.
My prediction is that this won't be possible until we are commonly implanting biometrics in people, but I guess we will wait and see!
[0] https://en.wikipedia.org/wiki/Korotkoff_sounds [1] http://www.visimobile.com/
Thanks for batting around the idea with me, and for the valuable references. I didn't know there was any other way to measure BP than the old-fashioned way.
http://www.racgp.org.au/download/documents/AFP/2011/November...
It's still not what you want, but it looks like things are moving in the direction you suggest. That said, I'm wondering if it is even possible to do what you suggest without being inconvenient to the user. Having an armband inflate and tighten around my arm every few seconds with become infuriatingly annoying. Are there alternatively ways to measure blood pressure that are imperceptible?
What I'm looking forward to seeing is conductive textiles making their way into compression clothing so we can measure heart rate all the time. i.e. a wearable EKG shirt. The use case would be older people at risk for a heart attack and heart fail so we can detect problematic heart abnormalities that are predictive of failure.
But now I see that the community actually downvoted your comment rather than rewarded it. Darn.
For what it's worth, and even though this reply is very late: Thank you so much for your time and for the thoughtful and helpful reply. The links, specifically, were the reason it was helpful to me.
We know that most people can't even take their medication properly (many organ transplants fail because people don't comply with the medication regime, for example) so easier blood pressure monitoring would probably be useful. Especially if you combine it with something that can lower blood pressure.
http://www.nice.org.uk/guidance/cg127/chapter/Key-priorities...
> Diagnosing hypertension
> If the clinic blood pressure is 140/90 mmHg or higher, offer ambulatory blood pressure monitoring (ABPM) to confirm the diagnosis of hypertension. [new 2011]
> When using ABPM to confirm a diagnosis of hypertension, ensure that at least two measurements per hour are taken during the person's usual waking hours (for example, between 08:00 and 22:00).
> Use the average value of at least 14 measurements taken during the person's usual waking hours to confirm a diagnosis of hypertension. [new 2011]
> When using home blood pressure monitoring (HBPM) to confirm a diagnosis of hypertension, ensure that:
> for each blood pressure recording, two consecutive measurements are taken, at least 1 minute apart and with the person seated and
> blood pressure is recorded twice daily, ideally in the morning and evening and
> blood pressure recording continues for at least 4 days, ideally for 7 days.
> Discard the measurements taken on the first day and use the average value of all the remaining measurements to confirm a diagnosis of hypertension. [new 2011]
Let me explain:
Use cases outlined in this article that have minimal real medical benefit, but sound great:
- Home blood testing (why? there are only 2 types of people who may find this useful: patients on Warfarin, and patients on dialysis. Dialysis patients (if not on home dialysis) get tested 3 times a week when they go to dialysis. INR is usually stable once established, ceteris paribus)
- Field blood testing (Mostly pointless. Clinical signs are much more useful in a casualty situation. Arterial blood gasses are useful in resus rooms in emergency departments but in the field you want to keep them breathing, not go reversing exotic blood gas abnormalities)
For the majority of tests (EUC, CMP, LFT, FBC, Coagulation studies) in rural/isolated places, the i-stat machine is used in australia. This is priced ok... around $40 per cartridge for around 6-10 results.
The i-stat works with not much more than a couple of drops of blood as well, although it needs to be drawn from a vein, not a finger-prick.
Other tests, like lipids, or antibody markers, etc, are rarer and much more expensive. It would be great to have them cheap (but again, we don't really know how cheap theranos is saying tests are going to be) which may make them more common, but again there just isn't much benefit in taking them regularly.
Theranos may well be a white elephant. which is unfortunate. The promised technology (fingerprint testing) would be great. But I can't comprehend how they could go for 12 years without a product. And as a doctor, I would be very suprised if the technology works as advertised. There is unfortunately too much hype surrounding this product for me to believe that we will get what has been promoted, but you never know and I would love to be pleasantly surprised.
I wonder how popular this viewpoint will be here on HN - the home of lifehacking and microanalysis of bodyfunction (often without the corresponding knowledge to go with it, but prefaced with a view that 'more data will always be better').