Can We Automate Our Way to Affordable Medicine?
By Wil Forbis
The United States has argued about the topic of health insurance for decades. In the early 2010s, when Obamacare was crafted and debated, several models for health insurance were thrown about, from single payer plans similar to those used in Canada and Europe, to health savings accounts. Eventually, Obamacare became the law of the land, but it now seems that status will be short lived. As I speak, the House has passed (and the Senate is deliberating) a new bill, the American Health Care Act, that shutters Obamacare.†
Itís possible that the challenges of health insurance may never be resolved. But it occurred to me recently that perhaps the need for health insurance can be bypassed. If medical care was simply much cheaper then we wouldnít need health insurance (though it would certainly still be nice to have.) †
How can we drive down medical costs? Weíve all observed that a number of non-medical goods have become cheaper over the years. For example: music. Consumers used to pay twelve bucks for a cd of ten songs and now we take for granted that millions of songs are available essentially for free on web sites like Youtube or Spotify. Or consider news and journalism. We used to plop down 20 bucks or more a month on magazines, whereas we can now find an infinite variety of writing on the web for free. †
Some might say these are apples and oranges comparisons, that we canít lower the cost of drugs and medical care the way we did music and other media. And they may be right. But letís consider how a product like like music lost its value. With the development of mp3s it was suddenly possible to duplicate and distribute music files to all ends of the earth. †The legality of this was debated, but it was largely a moot debate. Pirating music was so easy that meaningful prohibition was impossible. Could something similar happen with medical care? †
To explore that, we need to define what medical care is. Iíll break it down to three components: diagnostics, drugs and surgery. Thereís more to medicine than that, of course, but I think if we could lower the costs of these parts, weíd go a long way towards making health care more affordable. †
This is the process of a doctor observing a patientís symptoms, diagnosing a condition and prescribing an alleviating treatment like drugs or surgery. To accomplish this doctors train for several years and learn to identify the various diseases and conditions that manifest themselves in the human body. Can this diagnostic process be digitized?†
At least part of it can and, in fact, the process has started. Artificial intelligence tools like IBMís Watson are already ensconced in the medical establishment and, as this article at Medicalfuturist.com states, these tools have distinct advantages over flesh and blood doctors. †
The advantage Watson offers is an ability to comb through patient records, English textbooks, and millions of medical papers in existing databases ó in seconds instead of decades. Its algorithms arrive at diagnostic suggestions, and assign probable success rates to them. †
Itís worth noting that a recently developed diagnostic AI tool is better than human doctors at recognizing skin cancer. †
That said, we have a ways to go before we can defer to AI for all our diagnostics. The best human doctors have an intuitive sense that guides them to make difficult diagnoses. It may be years before AI can compete with those complex, subconscious process, if they every do. †
Nonetheless, AI doesnít need to be as good as human doctors to drive down prices. The beauty of AI is that once it has learned to successfully diagnose a certain condition (like skin cancer) that knowledge can be replicated in other instances of AI. Passing knowledge from one AI to a hundred is, at least theoretically, cheaper than creating a hundred new doctors. (Of course, because this process can be cheaper, doesnít mean it will be. The medical industry seems to have a magical ability to inflate costs at every turn.) †
Drugs, unlike music or diagnostic analysis, are material objects and thus impervious to digitizing. However, the real value of a drug is not the material product, but in the recipe. If you know how to make a drug and can access the raw materials you donít need a stockpile of 5000 pills.†
The recipes of patented drugsó-the formulationsó- are no secret; they are included in the patent. This recipe is basically just a set of instructions and thus could be digitized and propagated throughout the web. (This would probably be highly illegal and would occur mainly in what is known as the dark web.)
Armed with a recipe, how easy is it to manufacture drugs? Well, it varies by drug but we all know this sort of thing is possible. This is what manufacturers of illegal drugsóócocaine, heroin, methamphetamines, etc.ó-do all the time (with results of varying quality.) We also see counterfeit drugs manufactured in third world countries. These drugs are often of sub-par quality and can even be lethal. †
At this point Iím going to postulate a concept that may sound far fetched. Could a network of activists create a black market of ďpiratedĒ, affordable drugs? This should not be confused with the existing market of fake drugs run by smugglers and various criminals out to make a buck. Iím talking instead about people who would break patents to create functioning, counterfeit drugs and offer them at much lower costs than their legal counterparts? And could the threat of these black markets drive down legal drug prices? (In the same sense that the threat of pirate music sites enabled legal music distribution sites like Spotify to drive a hard bargain and offer miniscule payouts to music rights holders.) †
The work Yusuf Hamied might serve as an example of this sort of thing. He has been offering drug copies to various poor countries for years.
(Update, July 2018. The Four Thieves Vinegar Collective, profiled here, is another example.)
I should be clear here that Iím not advocating this. I do think patents and the financial payoff they engender drive innovation and development. But the world does not operate according to my whims; Iím trying to predict what will happen, not what should. †
The process of surgery is, of course, the most challenging to ďdigitize.Ē Surgery is a physical act (itís more a service than a product) and requires physical tools.†
Surgery now is dependent on the knowledge and skill housed in a surgeonís brain. Like all brain-contained knowledge, this is very difficult to duplicate and transfer. (Training surgeons takes years and is expensive.) †
There are a few tools on the horizon that could affect this dependency on brain knowledge. One is augmented reality and the second is robotic surgery. †
Most people are familiar with AR (augmented reality) glasses that overlay information on top of a userís view of the world. A basic example would be AR that places arrows over a personís field of vision, guiding them to a restaurant they are seeking. An advanced AR could guide a person through the process of assembling a bookshelf from parts. †
In theory, an AR tool could ďguideĒ a less knowledgeable (and thus cheaper) surgeon through a particular surgery. In effect, the information formally housed in the surgeonís brain would be housed in the AR software. That information, being digital, would thus be easily reproduced and shared to other AR devices.
(To be clear, Iím not saying any boob off the street could use AR to perform surgery. Iím merely saying surgeons would require less training and thus be cheaper.) †
A second tool that could come into play is robotic-assisted surgery. In this process, a human surgeon guides robotic tools that do the actual cutting. As a result, some surgery can be less invasive and more accurate. (The manual dexterity human surgeons need is often compared to that of a virtuoso musician.) In 2006, a robotic surgeon even performed surgery unassisted by human surgeons! †(Humans did monitor the procedure, of course.)
I think AR and robotic surgery hold some promise, but when they will affect the pricing structure of surgery is anyoneís guess. †
Summing it Up
Iíll restate my premise here. If we can digitize units of information or steps in a process, we can, in theory, lower their costs. Much of the medical establishment is information and process driven and susceptible to digitization.
That said, numerous factors hang in the air: regulation (or the willingness to ignore regulation), pushback from doctors, how effectively activists can make a moral case for this sort of thing, etc. †
I donít think automating our way to cheaper medicine will be easy, but I think itís possible, especially as technology progresses.
What do you think? Leave your comments on the Guestbook!
Wil Forbis is a well known international playboy who lives a fast paced life attending chic parties, performing feats of derring-do and making love to the world's most beautiful women. Together with his partner, Scrotum-Boy, he is making the world safe for democracy. Email - firstname.lastname@example.orgVisit Wil's web log, The Wil Forbis Blog, and receive complete enlightenment.