9 ms·

Good. I spent a few years working in a medical IoT startup, and the current state of the "mainstream" medical monitoring technology is not inspiring. During a

• hacker_9 9y ago
  The monitoring devices are the real problem, not the people. Do you think all weight spikes simply equate to having CHF? You can't just rely on such a basic measurement when dealing with an incredibly complex system such as the human body. A lot more specialist equipment is going to be needed to come to a final conclusion (Questioning from experts, MRIs, XRays, even Microsurgery with cameras). What consumers can currently access are weighing scales, and heartbeat monitors. Maybe even temperature monitors. Do you think this stuff is useful for diagnosing medical conditions? There is a whole lot going on under the surface that we simply can't see at all, that's the real problem.
  • rhombocombus 9y ago
    It is possible that some of the noise (non-medically relevant changes) could be filtered algorithmically, and conversely conditions might be able to be picked up that might not be noticed by a normal provider. Physicians use those tools to diagnose disease all the time, that's why there is a thermometer, scale, and stethoscope in every doctor's office in the world, I don't understand how these tools aren't useful.
    • hacker_9 9y ago
      They can tell you something is different, or out of the normal range, they don't give you the real cause of the problem though. This is left to the doctor who will take into account age, gender, medical history, as well as the fuzzy description of the problem from the patient. Then they'll give you some medication and hope that it heals in a few days, if not, then re-assess and do more trial and error diagnostics and medications (which can go on for months, even years). If we had technology that could monitor internal changes on a real time basis though, then the face of medicine would change forever.
  • killjoywashere 9y ago
    Your comment is a great example of the disconnect between people in this space and people not in this space. Do you actually think anyone in this space would believe that weight spikes are tied to CHF in all cases? You don't diagnose CHF by weight. It's something you trend in known patients. Comments like this are one reason why doctors give zero shits about the potential of computing. Explaining even the basics of a narrow area of human physiology to some overly arrogant IT guy is maddening (arrogant is fine, for Pete's sake, it's not like doctors aren't arrogant). Yes, the doctors can be blamed for conflating computer scientists with the IT guy who came to replace their mouse. But the IT guy and the CS guy (always a guy) can both equally be blamed for their unholy arrogance. How many lives did you save today? Zero? Ok, at least I didn't spill any data (that I know of). This can go round, and round, and round. You have no idea how far down the problem goes. I know this because no one knows how far the problem goes. We sequence hundreds of thousands if not millions of genomes, and still we don't know. Do you really think we spend a decade in training and come out thinking everything is simple?! Now, on the flip side, doctors also don't even know how to frame their problems, in no small part because they're only required to take 2 semesters of calculus. And then most of them punch out of math as fast as possible. Computer scientists should be going to biology conferences. Go to ASCO. Pick a medical specialty, they have multiple conferences a year, I garauntee it. There's one or 10 in your city. If you want middle ground, look at microbiology, immunology, molecular biology. They use a fair number of quantitative methods (sequencing, mass spec, flow cytometry, etc).
• SilasX 9y ago
  It seems like the real (meta)problem is that the system’s incentives don’t reward this. If a doctor’s take home pay increased as these things were avoided, you wouldn’t need some risk-tolerant startup to dive in with some innovative protocol; someone would have figured out a fix a long time ago. (Edit: to adapt the old saying: if life-saving were profitable, common sense would make us all saints. But alas, we live in a world where regulatory arbitrage pays more.) I think we should ask how we got to the point that the incentives are so incompatible and how we got to that point and how to realign them.
  • teej 9y ago
    Both doctors and hospitals are financially incentivized to avoid rehospitilization. It is actively tracked and directly impacts billing. Check out the "Hospital Readmissions Reduction Program" that Obamacare introduced. I’m sick of people spreading FUD about medical billing based simply on the ridiculous “the man is out to get me” narrative.
    • SilasX 9y ago
      Fair enough -- I was severely understating the extent to which there are financial incentives. But I think you're making a similar error in the opposite direction, coming from the approach of of "that problem is solved because we have a law for it." Imagine in any other context: "Everyone makes a living wage, because we have minimum wage laws." -> No, cost of living can eat it up in some areas, and some people are forced to go contractor, where take-home can get arbitrarily low. "Taxis provide vital, universal service because they're obligated to take you anywhere." -> Yeah, but in practice they can flout this with impunity. In this case, yes, the most obvious cases of hospital failure will be undercompensated, getting some incentives right. But is that the same thing as saying that actual, effective measures to avoid screwing up will translate into higher profits, and the people capable of identifying these will benefit from pointing them out? I think that's a stretch. Why do you think such a simple measure, as proposed by this startup, has to drag them kicking and screaming into implementation? Isn't this something a hundred doctors might have noticed and suggested? Something system level isn't working. Either the feedback loop between hospitals and doctors is broken, or it's too easy to relabel a readmission as being a new issue, or something like that. If startups have that kind of resistance to something so simple, that suggests a lot of easy improvements are still being held up by misaligned incentives, and supports my original point that system level fixes are a bigger target for health outcome returns.
• u801e 9y ago
  There are plenty of indicators that could be monitored in various patients for various diseases, but a lot of them may require doing things that are beyond the ability of someone who is not medically trained. Some of them require intravenous blood samples, for example. But others are like example you mention (which only requires that the patient weigh themselves at a given time each day). It would be nice if we could make it simpler for people to measure things like their own hemoglobin A1C, blood urea nitrogen/creatinine, blood flow around organs/extremities, etc. and send the data to a service that can be used to alert medical staff if something looks like it's getting worse.
• joshgel 9y ago
  Interesting you bring up heart failure readmissions. Not saying they aren't a problem, but here is a fascinating study in JAMA that argues that places with lower readmissions have higher (worse) mortality. https://jamanetwork.com/journals/jamacardiology/article-abstract/2663213 https://jamanetwork.com/journals/jamacardiology/article-abst... Not sure we (I'm a physician) understand even what the problems we need technology to solve are. But yes, we need technology to help (find and) solve some major problems that will improve lives.
  • blhack 9y ago
    Are there other examples of things you wish existed? Again, forget about if they could exist, just stuff you wish you had?
    • joshgel 9y ago
      Loads. A functional EHR (UX). An easy way to keep track of what medications a patient is one and has been on (?blockchain). Better quality and cost measures (big data).
  • obastani 9y ago
    Couldn't this finding be explained by the fact that the hospitals with lower readmissions tend to be poorer (since poorer patients are less likely to be able to afford care)? I only took a quick look, but I didn't see how the paper accounts for this and other possible causality issues. On the other hand, it's certainly plausible to me that lower readmissions can cause higher mortality.
    • joshgel 9y ago
      Sure. Or those hospitals gamed the system and didn't admit patients who should have been admitted. Or plenty of other explanations... We call readmissions rate a Surrogate Marker. Something that isn't a hard outcome that we care about (that much), but rather an indicator of quality. We need surrogates to assess quality because hard outcomes are rarer. But this paper makes us think that maybe we need better markers.
• SQL2219 9y ago
  The tech is the easy part. Integrating into a 24/7 high turn-over environment where all employees, IT included, are stretched to the max, is the real issue.
  • scruple 9y ago
    And pumping the data in to their myriad EHR systems, getting those records associated correctly across those systems+, the unintuitive UI/UX problems you'll inevitably face, as well as training issues. And God help you if it's a place where they home brew their own solutions instead of going with a vendor. My company is in the health care space, as well, and a lot of the problems _do_ look easy ("that's just a CRUD app, right?" easy) until it comes time to get them adopted in the health care space.
• NetOpWibby 9y ago
  Failure to market an IoT product is unfortunately my experience as well. I hope the medical industry gets better, re-hospitalization sounds ridiculous for something easy to detect.
• lazyasciiart 9y ago
  Wifi-enabled scales that connect to webservices already exist on amazon - I assume the hard part is getting some certification so doctors/nurses can 'prescribe' it to patients when they leave hospital or while they're in hospital and getting the webservice to talk to $hospitalSoftware, etc?
• ubernostrum 9y ago
  I'll throw in a slightly contrary opinion. I work for a health-insurance startup. And many of the people we cover are on extremely low fixed incomes. It's not unusual for them to be in situations where they have to choose which bill to pay this month and thus which utility will get shut off next month for non-payment. Their premiums are heavily subsidized and they only have insurance at all because of government programs, sometimes only because they're eligible for multiple programs which combine to help them out (and which are also constantly on the chopping block). So any kind of cellular phone, let alone a multi-hundred-dollar smartphone with service contract, is simply an unaffordable luxury for them. And that's why I worry about these kinds of efforts. More affluent people can benefit from smartphone-centric approaches, but a huge percentage of the US slips through the cracks when you do that. So solving America's health-care problem is not going to happen by focusing on people who have smartphones. That will just heighten the divide; the people who are already in a position to get good care will continue to get it, and the people who aren't will continue to get worse. If you want to avoid the grim-meathook-future problem, smartphone-centric (and, really, most tech-centric) approaches are unlikely to work.
• rayiner 9y ago
  > Can you imagine my frustration at hearing this? It's this massive problem with an obvious (CHEAP!! SO FUCKING CHEAP!!) solution. I suspect the nurse is there to perform the “getting the patient on the scale” function, not the “reading out the weight and tracking it” function.
  • delhanty 9y ago
    Perhaps the scales could be built into the bed somehow?
    • ams6110 9y ago
      Modern hospital beds do have scales built in.
      • delhanty 9y ago
        Thanks - I did not know that. Full-automation should be possible then - no need for the nurse to get the patient on the scales. That's probably what #rayiner meant when he wrote that the solution is obvious.
• ep103 9y ago
  Is there any chance you would be up for talking about where one could go about working in tech on problems like these? Every time I've looked, I usually just find hospitals maintaining poor IT infrastructure and CRM-type systems. I would love to work in this space. Strangely, I've been able to find startups in the education space, but not in healthcare. I imagine 90% of the problem would be integration into the behemoth that is medical administration, but that's something I also have the skillset for. Hacker news has a limit on replying to responses with less than a few votes, but my email is in my profile