Smartphone, finger prick, 15 minute diagnosis
February 5, 2015 10:20 AM Subscribe
Medical researchers at Columbia University have developed a smartphone accessory that can diagnose HIV and syphilis in 15 minutes. The device is estimated to cost $34 to manufacture, compared to $18,450 for existing equipment. It also does not require an external power source, which is a major benefit for use in areas without reliable access to electrical power.
Okay, if it's use is restricted to health care workers, fine. I wouldn't want the general public to be doing needle sticks.
posted by dances_with_sneetches at 10:35 AM on February 5, 2015
posted by dances_with_sneetches at 10:35 AM on February 5, 2015
Moore's law coming to medicine? Hope so.
posted by Ansible at 10:36 AM on February 5, 2015 [4 favorites]
posted by Ansible at 10:36 AM on February 5, 2015 [4 favorites]
Seriously though, someone should make a little tiny thermal cycler/fluorescent analyzer with small, pre-packaged inhibitor-tolerant assay kits for field qPCR/RT-qPRC testing (it would probably need its own battery to power the peltier, but should work in principle). Why hasn't anyone done this?
posted by Mitrovarr at 10:42 AM on February 5, 2015 [2 favorites]
posted by Mitrovarr at 10:42 AM on February 5, 2015 [2 favorites]
I wouldn't want the general public to be doing needle sticks.
It's not hard. Millions of diabetics do it every day, or several times per day. And there are a variety of kits on the market where you can take a blood sample yourself and then mail it to a lab for testing.
posted by Foosnark at 10:44 AM on February 5, 2015 [16 favorites]
It's not hard. Millions of diabetics do it every day, or several times per day. And there are a variety of kits on the market where you can take a blood sample yourself and then mail it to a lab for testing.
posted by Foosnark at 10:44 AM on February 5, 2015 [16 favorites]
Please correct me if I'm wrong, isn't there a window of time where you run the risk of false negatives for HIV testing due to how long it takes for antibodies to show up? The general public might not be ready for this.
posted by marimeko at 10:45 AM on February 5, 2015
posted by marimeko at 10:45 AM on February 5, 2015
It really would be great if things like this could do for medicine what Square has been able to do for credit card payments.
posted by maryr at 10:50 AM on February 5, 2015 [1 favorite]
posted by maryr at 10:50 AM on February 5, 2015 [1 favorite]
Please correct me if I'm wrong, isn't there a window of time where you run the risk of false negatives for HIV testing due to how long it takes for antibodies to show up?
Yup.
"Once a person is infected with HIV, it generally takes about 3 months for the body to produce enough antibodies to be detected by an HIV antibody test. (For some people, it can take up to 6 months.)"
posted by echocollate at 10:51 AM on February 5, 2015
Yup.
"Once a person is infected with HIV, it generally takes about 3 months for the body to produce enough antibodies to be detected by an HIV antibody test. (For some people, it can take up to 6 months.)"
posted by echocollate at 10:51 AM on February 5, 2015
OK, everyone who's "worried about the implications" - think about all those EPTs sold in pharmacies. AND YET society hasn't run itself aground.
This is a great advance. Period.
posted by IAmBroom at 10:55 AM on February 5, 2015 [22 favorites]
This is a great advance. Period.
posted by IAmBroom at 10:55 AM on February 5, 2015 [22 favorites]
Every day we're getting closer to Star Trek levels of medicine. I for one welcome our new alien ant overlords.
posted by Ruthless Bunny at 10:58 AM on February 5, 2015 [4 favorites]
posted by Ruthless Bunny at 10:58 AM on February 5, 2015 [4 favorites]
I'm not convinced that this approach has any advantages over a good lateral flow assay. Lateral flow assays don't need any power at all. They don't even need a smartphone. But if you want to integrate them with a smartphone-linked database/remote diagnostic service, you can just take a picture.
Plus, they're already on the market. Including tests for HIV and syphilis. People have been banging away at microfluidic diagnostics for 20 years, but the market continues to be dominated by lateral flow tests and dip strips. It's going to be hard to outdo cost of manufacture or ease of use of either of these.
As a tangent, I learned last week that the largest market for gold nanoparticles is manufacturers of lateral flow assays.
posted by mr_roboto at 10:58 AM on February 5, 2015 [2 favorites]
Plus, they're already on the market. Including tests for HIV and syphilis. People have been banging away at microfluidic diagnostics for 20 years, but the market continues to be dominated by lateral flow tests and dip strips. It's going to be hard to outdo cost of manufacture or ease of use of either of these.
As a tangent, I learned last week that the largest market for gold nanoparticles is manufacturers of lateral flow assays.
posted by mr_roboto at 10:58 AM on February 5, 2015 [2 favorites]
OK, everyone who's "worried about the implications" - think about all those EPTs sold in pharmacies. AND YET society hasn't run itself aground.
This is a great advance. Period.
The problem is that we can't adequately assess the methodology or testing because it's behind a paywall. This is sciencedaily.com, not a journal, and access to the journal article is restricted behind a fuckin' paywall like everything else. The only things I can see (w/o forking over cash) is a overly excited blurb w/ pretty insanely good specificity/sensitivity ratings.
All the articles I can find on this are breathless clickbait 'see how this one device changes medicine forever' as opposed to myself (or expert opinions) on how the clinical testing and device testing was performed. Don't get me wrong, I'd like for this to be a reality, but I'm much more cynical about whether or not this a realistic or even possibly useful device in the long (and peer-/FDA-tested) long run.
posted by kurosawa's pal at 10:59 AM on February 5, 2015 [2 favorites]
Sciencedaily sucks.
The term is "point of care sensor", this is nothing new; people have been talking about it properly since the late 90's (and implementing it since the mid 2000s).
e.g Here are >10,000 papers since 2011 on the same topic. If you have access, here is a review from 2014 with state of the art.
posted by lalochezia at 11:01 AM on February 5, 2015 [1 favorite]
The term is "point of care sensor", this is nothing new; people have been talking about it properly since the late 90's (and implementing it since the mid 2000s).
e.g Here are >10,000 papers since 2011 on the same topic. If you have access, here is a review from 2014 with state of the art.
posted by lalochezia at 11:01 AM on February 5, 2015 [1 favorite]
This seems like a very useful medical application. I do see why so many would be concerned with "needle sticks." Like one of the other commenters said, though, diabetics handle needles and insulin on a daily basis.
posted by Eaglemat at 11:11 AM on February 5, 2015
posted by Eaglemat at 11:11 AM on February 5, 2015
OK, everyone who's "worried about the implications" - think about all those EPTs sold in pharmacies. AND YET society hasn't run itself aground.
Women who get a false negative from EPT don't pose a risk to society. HIV-positive people who get a false negative from an HIV test absolutely do.
Even making the big assumption that there are no false negatives, a false positive rate of even 5% can be very problematic.
Can't help but notice that there's no information at all about accuracy in the article.
posted by neckro23 at 11:23 AM on February 5, 2015 [2 favorites]
Women who get a false negative from EPT don't pose a risk to society. HIV-positive people who get a false negative from an HIV test absolutely do.
Even making the big assumption that there are no false negatives, a false positive rate of even 5% can be very problematic.
Can't help but notice that there's no information at all about accuracy in the article.
posted by neckro23 at 11:23 AM on February 5, 2015 [2 favorites]
kurosawa's pal: Don't get me wrong, I'd like for this to be a reality, but I'm much more cynical about whether or not this a realistic or even possibly useful device in the long (and peer-/FDA-tested) long run.OK, now that's a realistic criticism: "Is this marketing bullshit?", not "What kind of horrors await a world where anyone can diagnose disease for $1?".
posted by IAmBroom at 11:29 AM on February 5, 2015 [2 favorites]
The solution to dealing with a new technology that might mislead people in certain cases is to educate people on its use, not to yell "but the apes are not ready for this dangerous new tool!" Sort of how the solution to the spread of HIV is to educate people on transmission and protection, not to outlaw sex and bleeding.
posted by Behemoth at 11:34 AM on February 5, 2015 [4 favorites]
posted by Behemoth at 11:34 AM on February 5, 2015 [4 favorites]
There are really good reasons that HIV testing through medical providers is almost always associated with counseling that have nothing to do with false positives/negatives.
posted by PMdixon at 11:38 AM on February 5, 2015 [1 favorite]
posted by PMdixon at 11:38 AM on February 5, 2015 [1 favorite]
"HIV-positive people who get a false negative from an HIV test absolutely do."
No, individuals who assume all their sexual- and drug-using partners are HIV negative do.
posted by docgonzo at 11:45 AM on February 5, 2015
No, individuals who assume all their sexual- and drug-using partners are HIV negative do.
posted by docgonzo at 11:45 AM on February 5, 2015
Neckro actually does have a good point about someone self-testing and getting a false negative and being lulled into a false sense of security. That's a fair point.
However, for me that is being countered by the point that "maybe this will catch the people who have been too afraid/embarrassed/busy/uninsured to go to a doctors' office for testing". I can't speak to whether one of those demographics is bigger than the other, but it's definitely a point I think is in this inventions' favor.
posted by EmpressCallipygos at 11:48 AM on February 5, 2015
However, for me that is being countered by the point that "maybe this will catch the people who have been too afraid/embarrassed/busy/uninsured to go to a doctors' office for testing". I can't speak to whether one of those demographics is bigger than the other, but it's definitely a point I think is in this inventions' favor.
posted by EmpressCallipygos at 11:48 AM on February 5, 2015
The problem is that we can't adequately assess the methodology or testing because it's behind a paywall.
Feel free to MeMail me if you want a copy of the article. STM is a perfectly respectable journal.
posted by no regrets, coyote at 11:58 AM on February 5, 2015
Feel free to MeMail me if you want a copy of the article. STM is a perfectly respectable journal.
posted by no regrets, coyote at 11:58 AM on February 5, 2015
Hopefully advances like this, especially in the realm of how inexpensive and portable they are, will have a big impact in third world countries. there will always be hand-wringing in places like the US, where the mere mention of a smartphone app leads to the assumption that it's created for use by teenage girls. carry on.
posted by OHenryPacey at 11:59 AM on February 5, 2015
posted by OHenryPacey at 11:59 AM on February 5, 2015
There are really good reasons that HIV testing through medical providers is almost always associated with counseling that have nothing to do with false positives/negatives.
Actually, counseling is no longer required, at least in California. HIV antibody testing is considered to be a preventive health test, like A1C or any other screening. You no longer need to sign a separate consent and you are tested automatically if you are pregnant.
posted by Sophie1 at 12:08 PM on February 5, 2015 [1 favorite]
Actually, counseling is no longer required, at least in California. HIV antibody testing is considered to be a preventive health test, like A1C or any other screening. You no longer need to sign a separate consent and you are tested automatically if you are pregnant.
posted by Sophie1 at 12:08 PM on February 5, 2015 [1 favorite]
Ansible: "Moore's law coming to medicine? Hope so."
Here's a Wired article from last year about someone who invented a way to do a whole bunch of tests on just a drop of blood (instead of, say, one vial of blood per test). It seems like it's a real company and it looks like they're already providing some kind of services in a few Walgreens, so I have to assume they've passed all the required regulatory stuff. Exciting times.
posted by mhum at 12:09 PM on February 5, 2015 [1 favorite]
Here's a Wired article from last year about someone who invented a way to do a whole bunch of tests on just a drop of blood (instead of, say, one vial of blood per test). It seems like it's a real company and it looks like they're already providing some kind of services in a few Walgreens, so I have to assume they've passed all the required regulatory stuff. Exciting times.
posted by mhum at 12:09 PM on February 5, 2015 [1 favorite]
Also, you know you can just go pick up an HIV test at the market today, or on Amazon by tomorrow, right? You have the answer immediately and it's saliva based.
This is nothing particularly new except that it involves a smartphone.
posted by Sophie1 at 12:12 PM on February 5, 2015 [2 favorites]
This is nothing particularly new except that it involves a smartphone.
posted by Sophie1 at 12:12 PM on February 5, 2015 [2 favorites]
Even more cost effective to R&D was the output of this young team of Ugandans who have figured out how to diagnose malaria using an app.
posted by infini at 12:12 PM on February 5, 2015 [4 favorites]
posted by infini at 12:12 PM on February 5, 2015 [4 favorites]
IF it is viable/reliable then really what needs to happen is that clear instructions on retesting timelines need to be implemented as well. Smartphone? Fine, once you test, it automatically sets an anonymous profile(X) and then sets your calendar for 3 & 6 month followups for (X), and clearly states one test does not mean you are clean.
posted by edgeways at 12:13 PM on February 5, 2015 [1 favorite]
posted by edgeways at 12:13 PM on February 5, 2015 [1 favorite]
Please correct me if I'm wrong, isn't there a window of time where you run the risk of false negatives for HIV testing due to how long it takes for antibodies to show up?
Yup.
There goes Saturday night. And my plans for in-app purchases to micro-monetize social platforms: TEN KREDZ TO TWEET SIGNED COPIES! FIFTEEN KREDZ FOR A MEDALLION ON GRINDR/BLENDR!!
Somebody get on that three-to-six month window, I'll go draw up some business plans and a Kickstarter.
posted by Ogre Lawless at 12:19 PM on February 5, 2015
Yup.
There goes Saturday night. And my plans for in-app purchases to micro-monetize social platforms: TEN KREDZ TO TWEET SIGNED COPIES! FIFTEEN KREDZ FOR A MEDALLION ON GRINDR/BLENDR!!
Somebody get on that three-to-six month window, I'll go draw up some business plans and a Kickstarter.
posted by Ogre Lawless at 12:19 PM on February 5, 2015
Even more cost effective to R&D was the output of this young team of Ugandans who have figured out how to diagnose malaria using an app.
An app and a device of some sort, according to that article. But it's still great to have a test with no disposables!
posted by maryr at 12:42 PM on February 5, 2015
An app and a device of some sort, according to that article. But it's still great to have a test with no disposables!
posted by maryr at 12:42 PM on February 5, 2015
For peeps interested in resource-limited molecular diagnostics, look into "isothermal amplification." It's like q/PCR but it can be done in a water bath or heat block without the need for thermocycling.
Add in an amplicon detection reagent like gold nonparticles (auNP) or hydroxynapthol blue and you can tell if the sample contains a specific sequence of DNA (ie., detection of a particular virus or even detection of viral subtype).
Say, for HPV, you can have one blood drop into several tubes or even better, use microfluidics to move around several drops of blood into each of the reaction tubes so you only need to get pricked once) which can tell you if you have the virus, and depending on which of the other tubes "light up" you can tell if you have a dangerous subtype or a more benign subtype.
Use a cellphone camera and an app to determine positivity/negativity against built in positive/negative controls.
Unfortunately, there are all kinds of problems with isothermal polymerases (some can amplify independent of template, some have a hard time accessing the DNA from un-highly-purified samples, among others, but the nonspecific activity is a pretty big one - leads to too many false positives unless you nerf the conditions and trade off a lot of sensitivity for some specificity).
posted by porpoise at 1:00 PM on February 5, 2015 [1 favorite]
Add in an amplicon detection reagent like gold nonparticles (auNP) or hydroxynapthol blue and you can tell if the sample contains a specific sequence of DNA (ie., detection of a particular virus or even detection of viral subtype).
Say, for HPV, you can have one blood drop into several tubes or even better, use microfluidics to move around several drops of blood into each of the reaction tubes so you only need to get pricked once) which can tell you if you have the virus, and depending on which of the other tubes "light up" you can tell if you have a dangerous subtype or a more benign subtype.
Use a cellphone camera and an app to determine positivity/negativity against built in positive/negative controls.
Unfortunately, there are all kinds of problems with isothermal polymerases (some can amplify independent of template, some have a hard time accessing the DNA from un-highly-purified samples, among others, but the nonspecific activity is a pretty big one - leads to too many false positives unless you nerf the conditions and trade off a lot of sensitivity for some specificity).
posted by porpoise at 1:00 PM on February 5, 2015 [1 favorite]
"Once a person is infected with HIV, it generally takes about 3 months for the body to produce enough antibodies to be detected by an HIV antibody test. (For some people, it can take up to 6 months.)"
This is a strong point of debate. Many of the modern testing mechanisms can detect HIV in a significantly shorter window period, although the FDA, et al. are afraid to come forward and definitively say so.
There's definitely a window period, but most people seem to agree that 6 months is extremely conservative.
However, the tests are certainly not perfect. These days, many HIV patients are essentially undetectable, because of how effective the modern treatments are (combined with how obnoxiously difficult it is to detect the virus in the first place).
Like PrEP, this seems like it might be an imperfect solution toward combating HIV. Education is still really, really, really important, but we should not let concern trolling prevent people (particularly those who are most at risk) from having access to tools that can dramatically lower their risk of infection, just because those tools aren't perfect.
posted by schmod at 1:25 PM on February 5, 2015 [1 favorite]
This is a strong point of debate. Many of the modern testing mechanisms can detect HIV in a significantly shorter window period, although the FDA, et al. are afraid to come forward and definitively say so.
There's definitely a window period, but most people seem to agree that 6 months is extremely conservative.
However, the tests are certainly not perfect. These days, many HIV patients are essentially undetectable, because of how effective the modern treatments are (combined with how obnoxiously difficult it is to detect the virus in the first place).
Like PrEP, this seems like it might be an imperfect solution toward combating HIV. Education is still really, really, really important, but we should not let concern trolling prevent people (particularly those who are most at risk) from having access to tools that can dramatically lower their risk of infection, just because those tools aren't perfect.
posted by schmod at 1:25 PM on February 5, 2015 [1 favorite]
Okay, if it's use is restricted to health care workers, fine. I wouldn't want the general public to be doing needle sticks.
Something like this could come close to eradicating these STDs. If I was still on the dating market and could pay $34 to have a fair chance of detecting if any potential dates had HIV, it would be an absolute game changer.
posted by corb at 1:32 PM on February 5, 2015
Something like this could come close to eradicating these STDs. If I was still on the dating market and could pay $34 to have a fair chance of detecting if any potential dates had HIV, it would be an absolute game changer.
posted by corb at 1:32 PM on February 5, 2015
If I was still on the dating market and could pay $34 to have a fair chance of detecting if any potential dates had HIV
* somewhere between 2 weeks and 6 months ago. Or has HIV and is being treated for it.
If you are in a high-risk dating pool, wear a condom, and take PrEP. Frequent testing is important, but definitely not adequate.
posted by schmod at 1:44 PM on February 5, 2015 [1 favorite]
* somewhere between 2 weeks and 6 months ago. Or has HIV and is being treated for it.
If you are in a high-risk dating pool, wear a condom, and take PrEP. Frequent testing is important, but definitely not adequate.
posted by schmod at 1:44 PM on February 5, 2015 [1 favorite]
a fair chance of detecting if any potential dates had HIV
Yeah, no, this isn't like "Thanks for the drinks, want to come up to my place for a blood test and some sex if it's negative?" All the negative result tells you is that someone probably hasn't had HIV for more than a few months. If they contracted the virus recently, or the test gives a false negative, they could certainly get a clean test despite having HIV.
posted by vytae at 2:09 PM on February 5, 2015 [1 favorite]
Yeah, no, this isn't like "Thanks for the drinks, want to come up to my place for a blood test and some sex if it's negative?" All the negative result tells you is that someone probably hasn't had HIV for more than a few months. If they contracted the virus recently, or the test gives a false negative, they could certainly get a clean test despite having HIV.
posted by vytae at 2:09 PM on February 5, 2015 [1 favorite]
vytae: Yeah, no, this isn't like "Thanks for the drinks, want to come up to my place for a blood test and some sex if it's negative?"
And if it was like that, it would probably still be a bad idea, because I can only imagine that most potential dates would run for the hills if you tried pulling that in the first few months of the relationship. STD testing is supposed to come later in the relationship, when you've been together a while and trust each other and want to stop using protection. Early on it's kind of pointless for the reasons mentioned above, and also because you can't yet trust your partner to be faithful (if you've ever declared yourselves to be exclusive yet). It's also an incredibly personal thing to ask.
posted by Mitrovarr at 2:17 PM on February 5, 2015
And if it was like that, it would probably still be a bad idea, because I can only imagine that most potential dates would run for the hills if you tried pulling that in the first few months of the relationship. STD testing is supposed to come later in the relationship, when you've been together a while and trust each other and want to stop using protection. Early on it's kind of pointless for the reasons mentioned above, and also because you can't yet trust your partner to be faithful (if you've ever declared yourselves to be exclusive yet). It's also an incredibly personal thing to ask.
posted by Mitrovarr at 2:17 PM on February 5, 2015
the negative result tells you is that someone probably hasn't had HIV for more than a few months. If they contracted the virus recently, or the test gives a false negative, they could certainly get a clean test despite having HIV.
Sure, it can give false negatives, and I certainly wouldn't go to protectionless sex on that basis, but it doesn't afaik give many false positives, so it would still be a useful weeding feature.
posted by corb at 2:21 PM on February 5, 2015
Sure, it can give false negatives, and I certainly wouldn't go to protectionless sex on that basis, but it doesn't afaik give many false positives, so it would still be a useful weeding feature.
posted by corb at 2:21 PM on February 5, 2015
This article in the New Yorker profiling a company (Theranos) and their CEO who is working to make basic blood tests vastly simpler and cheaper is pretty fascinating.
Theranos is working to make its testing available to several hospital systems and is in advanced discussions with the Cleveland Clinic. It has also opened centers in forty-one Walgreens pharmacies, with plans to open thousands more. If you show the pharmacist your I.D., your insurance card, and a doctor’s note, you can have your blood drawn right there. (The sample is then sent to a Theranos lab.) From that one sample, Holmes said, several tests can be run—all less expensive than standard blood tests, sometimes as much as ninety per cent below the rates that Medicare sets. A typical lab test for cholesterol can cost fifty dollars or more; the Theranos test at Walgreens costs two dollars and ninety-nine cents.posted by amanda at 2:45 PM on February 5, 2015 [2 favorites]
HIV testing is complicated (the most contagious people are newly-infected people inside the testing window who aren't being treated), but in terms of making sure people get adequate care as soon as possible, both for their sake and for others', and don't put off testing due to stigma or privacy worries, at-home testing is great. And as Sophie1 pointed out, it totally already exists (Oraquick).
posted by en forme de poire at 3:02 PM on February 5, 2015
posted by en forme de poire at 3:02 PM on February 5, 2015
My understanding is the smartphone dongle discussed here simultaneously tests for syphilis. This is important from both safer sex and treatment POVs because they tend to be co-morbid (40-50% among MSM?), and syphilis can be tested for within weeks after infection, rather than months as for HIV.
The more we know, the better decisions we can make. Not perfect, not simple, no doubt unintended consequences (forced testing on suspected sex workers, anyone?), but definitely more helpful than otherwise.
posted by Dreidl at 9:57 PM on February 5, 2015 [2 favorites]
The more we know, the better decisions we can make. Not perfect, not simple, no doubt unintended consequences (forced testing on suspected sex workers, anyone?), but definitely more helpful than otherwise.
posted by Dreidl at 9:57 PM on February 5, 2015 [2 favorites]
"No, I said prick your finger, not the other way around."
posted by Joe in Australia at 10:12 PM on February 5, 2015 [4 favorites]
posted by Joe in Australia at 10:12 PM on February 5, 2015 [4 favorites]
Also, the social aspect... at least in some circles I've run in, there's no playing with wet bits, not even safer sex with barriers, unless current test results are exchanged or everyone goes to the STD clinic together for testing. It's less HIV, and more HSV, Hep B & C and chlamydia the AFABs in my crowd are worried about.
After burying hundreds of my friends due to HIV, and living through Pelvic Inflammatory Disease myself from a youthful chlamydia infection, I think "fluid bonding" is a flagrant example of how stupid love can make people. No thanks. But I live in a large queer community in the affluent developed world, I'm not a sex worker and I'm fairly self-supporting financially; I can make the choice to only use barriers and still have a chance of keeping partners while staying physically safe and STD-free.
posted by Dreidl at 10:50 PM on February 5, 2015 [1 favorite]
After burying hundreds of my friends due to HIV, and living through Pelvic Inflammatory Disease myself from a youthful chlamydia infection, I think "fluid bonding" is a flagrant example of how stupid love can make people. No thanks. But I live in a large queer community in the affluent developed world, I'm not a sex worker and I'm fairly self-supporting financially; I can make the choice to only use barriers and still have a chance of keeping partners while staying physically safe and STD-free.
posted by Dreidl at 10:50 PM on February 5, 2015 [1 favorite]
« Older "Keep it 100!" | Pre-9/11 Ties Haunt Saudis as New Accusations... Newer »
This thread has been archived and is closed to new comments
posted by Mitrovarr at 10:23 AM on February 5, 2015 [2 favorites]