The Internet hasn’t notably cracked the health-care industry yet. It’s nibbled around the edges a little bit—for example, I can renew prescriptions online instead of phoning them in—but it hasn’t fundamentally changed the business model or the relationships between the players as in other industries. I’ll begin my fourth year as MSDN Magazine’s resident Diogenes (bit.ly/Xr3x) by predicting how it soon will.
The forces currently tearing apart the structure of higher education are also gathering in the medical industry. Health care in the United States consumes about $3 trillion per year, approximately one-sixth of the U.S. gross domestic product. The providers are partying like it’s 2006 and costs are spiraling. The population is aging and getting fatter; the boomer bulge is making its way through the demographic snake. Disruptive technologies are ready to rock. Unstoppable forces are slamming into immovable objects. Something is about to give.
Why hasn’t it happened yet? Partly because of the medical establishment’s famed inertia. Consider Ignaz Semmelweis, the 19th century Viennese obstetrician who lowered maternal mortality by 90 percent, simply by insisting that doctors wash their hands before examining childbirth patients. As reward for this spectacular improvement, his colleagues threw him into an insane asylum where he quickly died (see bit.ly/SO3jd4).
Perhaps faster technological change in medicine had to wait until kids who grew up with the Internet had finished medical school. That’s starting to happen now. A young doctor who attended a class I taught on Microsoft HealthVault told me: “My kids’ babysitter makes better use of the Internet in her business than we do here at [a major teaching hospital]. I’m here to learn how to fix that.” The tipping point where these guys accumulate enough power to change things is not far off.
Last month I explored how massive open online courses (MOOCs) are successful in education because they combine higher quality with lower cost. This virtuous combination is now approaching for the medical industry. Consider your child waking up saying, “I don’t feel good.” Instead of schlepping to the doctor’s office, suppose you could talk to a nurse on a Skype video link.
The world’s finest doctors would work out the diagnostic protocols for a sick kid, and a software wizard would walk the nurse through it. She would have trained extensively on this specific scenario, using excellent simulators, so she’d be an expert on it. The programs would continuously update the diagnostic probabilities based on the latest results seen in the local area, making her more current and precise than an unaided pediatrician today. Yet her time would cost far less than that of an MD.
What’s more, you would own a small instrument to measure and transmit your child’s temperature, blood pressure, pulse oxygenation and other vital signs. A camera on the instrument would transmit pictures of the throat or ear canal or skin rashes, and a microphone would transmit breath sounds and heartbeats. All this is under development, with prototypes already emerging and first commercial releases within the year (see econ.st/X5mq3e). Perhaps algorithms could compare the pictures and sounds you transmit to every other captured sample. Doctors wouldn’t misdiagnose diseases such as measles (which is rare in the United States) because they had never seen a case.
Patients with viral infections would be told to stay home, keep warm, take Tylenol and call back if they didn’t improve. Patients with bacterial infections or more severe symptoms would have prescriptions transmitted to a pharmacy for delivery that day. You wouldn’t have to take the whole day off from work to drive your kid to the doctor and exchange germs with everyone in the waiting room.
Patients with more serious conditions, or conditions that can’t be evaluated over the wire, would get appointments with the doctor that afternoon. Each doctor’s time would be far better utilized as well—mornings for follow-ups, afternoons for new cases from the Web nurses.
Just as the education industry will still need mentor classes for advanced topics, the medical industry will always need specialists and surgeons for when people get really sick. But the mass of day-to-day grunt work will be automated faster than anyone imagines, in the same way and for the same reasons as the teaching of freshman calculus is being transformed today.
The medical industry’s dam hasn’t yet cracked the way the education industry’s has. But there’s far more force building up behind the medical dam. The burst will be all the more spectacular when it comes, with concomitantly larger profit opportunities for developers and companies who are thinking forward. Call me if you’d like to discuss it.
David S. Platt teaches programming .NET at Harvard University Extension School and at companies all over the world. He’s the author of 11 programming books, including “Why Software Sucks” (Addison-Wesley Professional, 2006) and “Introducing Microsoft .NET” (Microsoft Press, 2002). Microsoft named him a Software Legend in 2002. He wonders whether he should tape down two of his daughter’s fingers so she learns how to count in octal. You can contact him at rollthunder.com.
As an EDI manager in Health care that are so many regulations for privacy around health care data, that moving to the internet brings liablity issues to companies. There are good private health software for itegration that is moving this way, however until we can resolve the liablity issues most companies will not risk it. In some cases there is medcial data on mental health, AIDS, VD that could be disasterous to people if seen. We talk about data security and data vaults, but these are only as good as they are until they are cracked, and the $$ value to crack these make it a lucrative challange, and they will all have a vulnerabilty somewhere. More the cost is in the medical supplies and drugs then the actual cost of doctor visits. The government regulates the medical practioners, but what about the cost of the other products?
Dear Dr. Smith: You are right about the licensing and reimbursement problems under the current system. However, I foresee a shift of licensing authority from state to Federal level as the latter starts assuming more of the payment burden. For example, to work for the Veterans Administration, which is run directly by the Feds, a doctor today needs only a valid license from any state, not one from the state in which that VA hospital is located, because the VA jurisdiction is Federal. Further, I foresee the shift from fee-for-service to some sort of capitation as inevitable under the current regime of cost increases, in which case the savings of diverting a routine problem to the cheaper NP accrue to the provider. That’ll make it happen. As to your assertions that EMR systems reduce MD productivity, I’ve seen it myself and heard MDs and others howling about it. It will eventually improve, I surely do hope, when developers start putting themselves into their users’ shoes. My Master Class on User Experience, in Boston this May (see www.idesign.net) addresses these sorts of issues. I do not have the knowledge to dispute you on the balance of art vs. science in medical diagnosis. I will say that technology -- if and when it’s done right, which it isn’t yet -- has the capability to handle the scientific portion of it, which humans aren’t great at, thereby freeing the humans to concentrate on the art portion of it, which only humans can do. For an example, find a video of my Why Software Sucks keynote talk somewhere, and see my example of auto-correct of spelling in Word. It uses the computer for what computers are good at (diligent pattern matching), thereby freeing the user, the writer, to conceive and express ideas, which is what humans are good at. It understands, it respects, it even enhances the humanity of the user. THAT is what software can do for people. Does medical software have a long way to go before we get there? Heck yes, a very long one. But the cost pressure is immense, something WILL give. Thanx for reading and for writing.
I realize you only have one page to write this article so you cannot possibly list all of the consequences to "digitizing" medical services. However, it seems that you portray this migration as all butterflies and roses. There are inherent risks that this type of service introduces. Yes, it would be a good money maker for the software developer and it may make the transfer of data more efficient. Yet, the largest concern I have is that the greatest risk we lose is the relationship between doctor and patient. A good example of this is my mother. She had schizophrenia and bipolar disorder. The psychiatric/counseling facility that rendered her services got rid of her psychiatrist that she had built a relationship with. This was to save the facility money. They went to the very program you are mentioning and signed her up w/ a psychiatrist in Tennessee. The cost of doing this then is that a relationship could not be built with a T.V. set. My mother committed suicide shortly thereafter. Had the doctor been local and available on a hand-to-hand basis, they could've sensed the warning signs. Instead, cost savings for them created a relationship that was syntheticly digitized for her. Likewise, going this route did not save my mom any money. The facility still billed the same amount and that is what we will see. Doctor offices in my area charge $150/visit when they only meet w/ people on average for 15-30 minutes. That's $300-$600 /hour of earnings. This type of tool will save the patient nothing but benefit the software developer and the doctor.
There are several barriers to the telephone nurse medical service model. First, there are state licensing requirements that limit what nurses can legally perform. Second, experience with this approach has identified a number of limitations in the accuracy of diagnostic conclusions. “Nurse Help Lines” have been around for 20 years but have not resulted in any cost savings. The service adds another layer of cost to the health care process. Third, remote medical services are not reimbursed to the provider of the service. Current medical insurance limits payment to physicians or providers of health services when the service is provided by telephone or telemedicine. There are billing codes for these remote services but most insurers will not recognize them for payment. As to the use of structured algorithms to enhance the accuracy of medical diagnosis and treatment, the physicians generally reject the process as time consuming and overly structured. Physicians make diagnostic and treatment decisions based on pattern recognition (combinations of symptoms, appearance and history of other medical conditions). Unfortunately, they make their decisions very quickly and tend to reject observations or findings that are inconsistent with their immediate conclusions. This decision process is clearly flawed but it is also very efficient and productive. I have previously developed structured protocols for the full range of emergency department medical presentations. Based on my experience with this effort (product was widely adopted) was that it required physicians to enter more data than they felt they needed to acquire. It also reduced productivity – patients that could be treated per hour of their time. Current electronic medical records reduce physician productivity by a minimum of 10% and up to 40% for the less efficient versions. There is a huge challenge to develop a “safe and practical” clinical database for the wide variety of symptoms that present to physicians for evaluation and treatment. The obvious trade off is how much data is required to support a robust diagnostic and treatment algorithm compared with the impact of the time to collect and record the data. There is also the important issue of errors in diagnosis and treatment associated with gathering less information to gain acceptance of the process. Like it or not, the evaluation of the sick patient includes a great deal of “art” and less science than most people believe. In my opinion, any successful solution to this dilemma will have to be focused on the patient’s role in providing more information and improving the physician’s starting data set.
I sincerely cannot wait until online health diagnostics are outsourced to some distant country where <your primary language here> is a second language or dialecticly incomprehensible. That has worked so well for customer support in other "service" industries.
I think we’re now at the point in putting medical data online that we were at with retail around 1997 and with finance around 2002. We wanted the benefits of online operations, but worried about the bad things that might happen to us. So we dipped a toe in the water (ordered one CD online, tried one financial transaction), found out that the benefits outweighed the costs, and kept on going. We’re just now dipping that toe in the water for medical data. As the benefits become more concrete, we’ll flock to it very quickly, as we did with everything else. The dam is just starting to crack, but those cracks will spread all the more quickly because of the brittleness of the current systems (or lack thereof).
I'm a little shocked by the resistance on the medical community to open up more. We've been transmitting our financial information - which is arguably more damaging ending up in the wrong hands than medical info - for at least a decade. CIGNA also tried (and dropped within about 6 months) the MD Live program which allowed you to see a doctor and send video cam/pictures and get diagnosis over the internet. We never got a chance to use it because they cancelled it, but it would be nice to have with younger children in the house. My pain point that technology can solve is answering the same damn questions at every provider, ad nauseum. I'm all for bio tech solving this, but for some reason or another it's nowhere to be found... so we all sit filling out 5 pages of medical history on a clipboard no matter where we go.
Competition is often proposed as a solution to rising medical costs. In a few limited areas, generally those not covered by insurance, it has worked well. For example, competition in vision correction surgery has drastically lowered cost and raised quality. However, much of medical care is purchased on an acute basis that doesn’t allow consumer choice to function. For example, “Help, I’m having a heart attack! Who wants to bid on taking care of me? I’ll take the lowest.” In general I'm in favor of competition, but this one's tricky.
Do you see any potentials for the Internet to drive medical costs down, not only by improving diagnostic efficiency, but particularly by enhancing competition within the medical industry?
More MSDN Magazine Blog entries >
Browse All MSDN Magazines
Subscribe to MSDN Flash newsletter
Receive the MSDN Flash e-mail newsletter every other week, with news and information personalized to your interests and areas of focus.